建立互動¶
您目前檢視的是 Interactions API 的 beta 版本。端點位於 /v1beta/ 之下。另有穩定的 v1 版本 可供使用。
POST https://generativelanguage.googleapis.com/v1beta/interactions
建立新的互動。
請求主體¶
**model**:`ModelOption` 用於產生互動的 `Model` 名稱。**未提供 `agent` 時為必填。**
可能的值: - `gemma-4-26b-a4b-it`:Gemma 4 26B A4B IT - `gemma-4-31b-it`:Gemma 4 31B IT - `gemini-flash-latest`:Gemini Flash 的最新版本 - `gemini-flash-lite-latest`:Gemini Flash-Lite 的最新版本 - `gemini-pro-latest`:Gemini Pro 的最新版本 - `gemini-2.5-flash`:我們的第一個混合推理模型,支援 100 萬個權杖的上下文視窗,並具備思考預算。 - `gemini-2.5-pro`:我們最先進的多用途模型,擅長程式設計與複雜推理任務。 - `gemini-2.5-flash-lite`:我們最小、最具成本效益的模型,為大規模使用而打造。 - `gemini-3.5-flash-lite`:Gemini 3.5 Flash-Lite——我們最快速、最具成本效益的 3.5 模型,適合高輸送量執行。 - `gemini-2.5-flash-image`:我們的原生圖片生成模型,針對速度、彈性與上下文理解最佳化。文字輸入與輸出的計價與 2.5 Flash 相同。 - `gemini-3-flash-preview`:我們為速度打造的最聰明模型,結合前沿智慧與卓越的搜尋及接地能力。 - `gemini-3.1-pro-preview`:我們最新的 SOTA 推理模型,具備前所未有的深度與細膩度,以及強大的多模態理解與程式設計能力。 - `gemini-3.1-pro-preview-customtools`:Gemini 3.1 Pro Preview,針對自訂工具使用最佳化 - `gemini-3.1-flash-lite`:我們最具成本效益的模型,針對大量代理任務、翻譯與簡單資料處理最佳化。 - `gemini-3.1-flash-lite-image`:Gemini 3.1 Flash Lite Image。 - `gemini-3-pro-image`:Gemini 3 Pro Image - `nano-banana-pro-preview`:Gemini 3 Pro Image Preview - `gemini-3.1-flash-image`:Gemini 3.1 Flash Image。 - `gemini-nano-banana-2.1`:Gemini Nano Banana 2.1。 - `gemini-3.1-flash-tts-preview`:Gemini 3.1 Flash TTS:強大、低延遲的語音生成。享有自然的輸出、可引導的提示詞,以及全新的富表現力音訊標籤,精確掌控旁白。 - `gemini-3.5-flash`:Gemini 3.5 Flash——我們較早的 Flash 模型,為日常、高輸送量工作負載的速度與基礎效能而打造。 - `gemini-3.6-flash`:Gemini 3.6 Flash——我們前一代的 Flash 模型,在一般代理與日常任務中兼顧速度與多模態能力。 - `gemini-3.7-flash`:Gemini 3.7 Flash——我們高速、高效率的 Flash 模型,為日常程式設計、代理工具使用與可靠的多步驟執行而打造。 - `gemini-3.8-flash`:Gemini 3.8 Flash——我們最聰明的 Flash 模型,為長程軟體工程、自主代理與複雜的企業工作流程而設計。 - `gemini-4-argon`:Gemini 4 Argon——我們全新的前沿模型,為以深度與精確度解決最困難的問題而打造,涵蓋真實世界的軟體工程與知識工作。 - `gemini-4-carbon`:Gemini 4 Carbon——我們能力最強的推理模型,為持續的代理程式設計、多模態知識工作、科學發現與網路安全防禦而打造。 - `gemini-4-hydrogen`:Gemini 4 Hydrogen——具備前沿級智慧的模型,為程式設計、代理工作流程與多模態知識工作的快速、可靠執行而打造。 - `gemini-3.8-flash-tts`:Gemini 3.8 Flash TTS——旗艦 TTS 模型,支援 Voice Design 與雙講者劇本控制。可提示自訂聲音角色、逐行指導語音呈現,並加入聲音爆發效果。 - `gemini-3.8-flash-lite-tts`:Gemini 3.8 Flash Lite TTS——高速且具成本效益,適合快速配音、媒體在地化與高輸送量語音代理。可直接取代 gemini-3.1-flash-tts-preview。 - `gemini-3.5-transcribe`:Gemini 3.5 Transcribe——低延遲的語音轉文字模型,具備以語句為單位的語言偵測、講者分離、單字級時間戳記與自訂詞彙偏置。 - `lyria-3-clip-preview`:我們的低延遲音樂生成模型,針對高傳真音訊片段與精確的節奏控制最佳化。 - `lyria-3-pro-preview`:我們先進的完整歌曲生成模型,具備深厚的作曲理解力,針對精確的結構控制與多樣音樂風格間的複雜轉場最佳化。 - `gemini-robotics-er-1.6-preview`:Gemini Robotics-ER 1.6 Preview - `gemini-robotics-er-2-preview`:Gemini Robotics Embodied Reasoning 2 Preview - `lyria-3.5`:我們的旗艦音樂生成模型,針對具複雜結構連貫性的完整長度歌曲最佳化。 - `gemini-omni-1.2-flash`:Gemini Omni 1.2 Flash——我們最新的多模態模型,支援快速、對話式的影片生成、編輯與電影級控制。 - `gemini-omni-1.1-flash`:Gemini Omni 1.1 Flash——我們的高效能多模態模型,專為快速、對話式的影片生成、編輯與電影級控制而設計。 - `gemini-omni-flash-preview`:我們多模態模型的預覽版,支援對話式影片生成、編輯與電影級控制。**agent**:`AgentOption` 用於產生互動的 `Agent` 名稱。**未提供 `model` 時為必填。**
可能的值: - `deep-research-preview-04-2026`:Gemini Deep Research Agent - `deep-research-max-preview-04-2026`:Gemini Deep Research Max Agent - `antigravity-preview-05-2026`:使用 Antigravity 受管理代理執行需要推理、檔案操作與工具使用的多步驟任務。 - `antigravity-preview-09-2026`:使用 Antigravity 受管理代理執行需要推理、檔案操作與工具使用的多步驟任務。input:必填 Content 或 Content array 或 Step array 或 string
互動的輸入(Model 與 Agent 共用)。
**generation_config**:`GenerationConfig` **模型設定** 模型互動的設定參數。*為 `agent_config` 的替代選項。僅在設定 `model` 時適用。*
**seed**:`integer` 解碼時使用的種子,以確保可重現性。 **stop_sequences**:`string array` 會停止輸出互動的字元序列清單。**tool_choice**:[`ToolChoiceConfig`](/api-reference/interactions-api/create-interaction/#Resource:ToolChoiceConfig) 或 `string enum` 工具選擇設定。
**可能的類型**:**[ToolChoiceConfig](/api-reference/interactions-api/create-interaction/#Resource:ToolChoiceConfig)**
包含允許工具的工具選擇設定。**allowed_tools**:`AllowedTools` 允許的工具。
**mode**:`string enum` 工具選擇的模式。
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。**string enum**
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。**thinking_level**:`ThinkingLevel` 模型應產生的思考權杖層級。
可能的值: - `minimal`:幾乎不進行思考。 - `low`:低思考層級。 - `medium`:中思考層級。 - `high`:高思考層級。 - `xhigh`:超高思考層級,高於 `high`。**thinking_summaries**:`ThinkingSummaries` 是否在回應中包含思考摘要。
可能的值: - `auto`:自動思考摘要。 - `none`:不含思考摘要。**video_config**:`VideoConfig` 影片生成設定。
**task**:`string enum` 影片生成的選用任務模式。若未指定,模型會根據提供的文字提示詞與輸入媒體自動決定適當的模式。
可能的值: - `text_to_video`:僅從文字提示詞生成影片。 - `image_to_video`:從一或兩張來源圖片生成影片。第一張圖片定義起始影格,選用的第二張圖片定義結束影格。 - `reference_to_video`:使用參考媒體(例如圖片、音訊或影片)生成影片。 - `edit`:修改既有的輸入影片。 - `extend`:延伸既有的輸入影片。**transcription_config**:`TranscriptionConfig` 選用。語音辨識(轉錄)設定。若提供此設定,將啟用 ASR。
**language_codes**:`string array` 選用。BCP-47 語言代碼,提供音訊中所含語言的提示。若省略或留空,預設為自動語言偵測。 **custom_vocabulary**:`string array` 選用。自訂詞彙片語清單,用於引導語音辨識模型偏向辨識特定詞彙。**mode**:`TranscriptionMode` 或 `string enum` 具判別性的轉錄模式選項或列舉。
**可能的類型**:**SmartTranscriptionMode**
智慧轉錄模式的設定。 **type**:必填 `object` 值:`smart` **source_text**:`string` 要透過語音指令編輯或改寫的來源文字。提供時,模型會執行語音驅動的編輯,並輸出最終定稿的替代文字。**VerbatimTranscriptionMode**
逐字轉錄模式的設定。 **type**:必填 `object` 值:`verbatim` **timestamp_granularities**:`string array` 選用。轉錄輸出中要包含的時間戳記粒度。支援的值:"word"。若留空,則不產生時間戳記。 **diarization_mode**:`string` 選用。設定講者分離。支援的值:"speaker"。**string enum**
可能的值: - `verbatim`:逐字轉錄模式。 - `smart`:智慧轉錄模式。**structured_speech_config**:`SpeakerConfig` 或 `SpeechConfig array` 語音與多講者設定。
**可能的類型**:**SpeakerConfig**
多講者與語音生成的設定。**speakers**:`SpeechConfig array` 個別講者的設定。
**voice**:`string` 講者的聲音。 **language**:`string` 語音的語言。 **speaker**:`string` 講者的名稱,應與提示詞中給出的講者名稱相符。**SpeechConfig array**
語音互動的設定。 **voice**:`string` 講者的聲音。 **language**:`string` 語音的語言。 **speaker**:`string` 講者的名稱,應與提示詞中給出的講者名稱相符。**agent_config**:`AntigravityAgentConfig` 或 `DeepResearchAgentConfig` 或 `DynamicAgentConfig` **代理設定** 代理的設定。*為 `generation_config` 的替代選項。僅在設定 `agent` 時適用。*
**可能的類型**:**AntigravityAgentConfig**
Antigravity 代理執行階段的設定。提供伺服器端對代理執行環境與工具設定的控制。 **type**:必填 `object` 值:`antigravity` **model**:`string` 用於代理推理的模型。 **max_total_tokens**:`string` 代理執行的權杖總數上限。**DeepResearchAgentConfig**
Deep Research 代理的設定。 **type**:必填 `object` 值:`deep-research`**thinking_summaries**:`ThinkingSummaries` 是否在回應中包含思考摘要。
可能的值: - `auto`:自動思考摘要。 - `none`:不含思考摘要。**visualization**:`string enum` 是否在回應中包含視覺化內容。
可能的值: - `off`:不包含視覺化內容。 - `auto`:自動包含視覺化內容。**DynamicAgentConfig**
動態代理的設定。 **type**:必填 `object` 值:`dynamic`system_instruction:string
互動的系統指令。
tools:Tool array
模型在互動期間可呼叫的工具宣告清單。
**response_format**:[`ResponseFormat`](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) 或 [`ResponseFormat`](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) `array` 強制產生的回應為符合此欄位所指定 JSON 架構的 JSON 物件。
**可能的類型**:**[ResponseFormat](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat)**
**[ResponseFormat](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) array**
每個項目為下列其中之一的清單:`AudioResponseFormat`、`ImageResponseFormat`、`TextResponseFormat`、`VideoResponseFormat`。previous_interaction_id:string
前一次互動的 ID(如有)。
**environment**:[`EnvironmentConfig`](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentConfig) 或 `string` 互動的環境設定。可以是指定遠端環境來源的物件,或是參照既有環境 ID 的字串。
**可能的類型**:**[EnvironmentConfig](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentConfig)**
自訂環境的設定。 **type**:必填 `object` 值:`remote`**sources**:`Source array`
**type**:`string enum`
可能的值: - `gcs`:Cloud Storage 值區。 - `inline`:行內內容。 - `repository`:一般儲存庫。來源 URL 中的通訊協定前綴可識別提供者(例如 github://、gcs://)。**network**:[`EnvironmentNetworkEgressAllowlist`](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist) 或 `string enum` 環境的網路設定。
**可能的類型**:**[EnvironmentNetworkEgressAllowlist](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist)**
環境的網路輸出設定。**allowlist**:`EgressRule array` 或 `string enum` 允許的網域及其設定清單。設為 `"disabled"` 可封鎖所有網路輸出。
**可能的類型**:**EgressRule array**
單一網域允許清單規則,可選用標頭注入。 **domain**:必填 `string` 允許向外請求的網域。支援萬用字元(例如 '*.googleapis.com')。使用 '*' 可允許所有網域。 **transform**:`object` 要注入符合此規則之請求的標頭。鍵:標頭名稱(例如 "Authorization")。值:標頭值(例如 "Bearer your-token")。 **credential**:`string` 選用。以 ID 參照伺服器管理的 Credential 資源。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。**env**:`object` 或 `string` 要在沙箱環境中設定的環境變數。
**可能的類型**:**string**
直接的字串值。**mode**:`string enum` 此環境的運算生命週期模式。
可能的值: - `on_demand`:此環境的運算生命週期模式。 - `self_hosted`:此環境的運算生命週期模式。**string**
直接的字串值。**service_tier**:`ServiceTier` 互動的服務層級。
可能的值: - `flex`:Flex 服務層級。 - `standard`:Standard 服務層級。 - `priority`:Priority 服務層級。**webhook_config**:`WebhookConfig` 選用。用於在互動完成時接收通知的 webhook 設定。
**uris**:`string array` 選用。若設定,這些 webhook URI 將用於 webhook 事件,以取代已註冊的 webhook。 **user_metadata**:`object` 選用。每次向 webhook 發送事件時會一併傳回的使用者中繼資料。**safety_settings**:`SafetySetting array` 互動的安全設定。
**type**:必填 `HarmCategory` 要封鎖的危害類別類型。
可能的值: - `hate_speech`:基於特定屬性宣揚暴力或煽動對個人或群體仇恨的內容。 - `dangerous_content`:宣揚、助長或促成危險活動的內容。 - `harassment`:辱罵、威脅,或意圖霸凌、折磨或嘲弄的內容。 - `sexually_explicit`:含有露骨性內容的內容。 - `jailbreak`:旨在繞過安全篩選器的提示詞。**threshold**:必填 `string enum` 封鎖內容的門檻。若危害機率超過此門檻,內容將被封鎖。
可能的值: - `block_low_and_above`:封鎖危害機率為低或更高的內容。 - `block_medium_and_above`:封鎖危害機率為中或更高的內容。 - `block_only_high`:僅封鎖危害機率為高的內容。 - `off`:完全關閉安全篩選器。**method**:`string enum` 選用。封鎖內容的方法。若未指定,預設行為是使用機率分數。
可能的值: - `severity`:危害封鎖方法同時使用機率與嚴重程度分數。 - `probability`:危害封鎖方法使用機率分數。labels:object
帶有使用者自訂中繼資料的請求標籤。標籤鍵與值的長度不得超過 63 個字元(Unicode 碼位),且只能包含小寫字母、數字字元、底線與連字號。允許國際字元。標籤值為選用。標籤鍵必須以字母開頭。
continuation_token:string
用於恢復長時間解碼的不透明權杖。輸出:當狀態為 REQUIRES_CONTINUATION 且可恢復解碼時設定。輸入:在 CreateInteraction 中原封不動地傳回最新權杖,以繼續解碼。
stream:boolean
僅供輸入。互動是否以串流方式傳輸。
store:boolean
僅供輸入。是否儲存回應與請求以供日後擷取。
background:boolean
僅供輸入。是否在背景執行模型互動。
範例請求
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"input": "Hello, how are you?"
}'
from google import genai
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Hello, how are you?",
)
print(interaction.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
input: 'Hello, how are you?',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.input(InteractionsInput.of("Hello, how are you?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Hello, how are you?")),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"input": [
{ "type": "user_input", "content": [{ "type": "text", "text": "Hello!" }] },
{ "type": "model_output", "content": [{ "type": "text", "text": "Hi there! How can I help you today?" }] },
{ "type": "user_input", "content": [{ "type": "text", "text": "What is the capital of France?" }] }
]
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
input=[
{
"type": "user_input",
"content": [{"type": "text", "text": "Hello!"}],
},
{
"type": "model_output",
"content": [{
"type": "text",
"text": "Hi there! How can I help you today?",
}],
},
{
"type": "user_input",
"content": [
{"type": "text", "text": "What is the capital of France?"}
],
},
],
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
input: [
{type: 'user_input', content: [{type: 'text', text: 'Hello'}]},
{
type: 'model_output',
content: [
{type: 'text', text: 'Hi there! How can I help you today?'},
],
},
{
type: 'user_input',
content: [{type: 'text', text: 'What is the capital of France?'}],
},
],
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.ModelOutputStep;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.interactions.TextContent;
import com.google.genai.gaos.models.interactions.UserInputStep;
Client client = new Client();
List<Step> conversation =
List.of(
UserInputStep.builder()
.content(List.of(TextContent.builder().text("Hello!").build()))
.build(),
ModelOutputStep.builder()
.content(
List.of(
TextContent.builder().text("Hi there! How can I help you today?").build()))
.build(),
UserInputStep.builder()
.content(
List.of(TextContent.builder().text("What is the capital of France?").build()))
.build());
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.input(InteractionsInput.ofStep(conversation))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
history := []gaos_interactions.Step{
gaos_interactions.NewStep(gaos_interactions.UserInputStep{
Content: []gaos_interactions.Content{
gaos_interactions.NewContent(gaos_interactions.TextContent{
Text: "Hello!",
}),
},
}),
gaos_interactions.NewStep(gaos_interactions.ModelOutputStep{
Content: []gaos_interactions.Content{
gaos_interactions.NewContent(gaos_interactions.TextContent{
Text: "Hi there! How can I help you today?",
}),
},
}),
gaos_interactions.NewStep(gaos_interactions.UserInputStep{
Content: []gaos_interactions.Content{
gaos_interactions.NewContent(gaos_interactions.TextContent{
Text: "What is the capital of France?",
}),
},
}),
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput(history)),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"input": [
{
"type": "text",
"text": "What is in this picture?"
},
{
"type": "image",
"data": "BASE64_ENCODED_IMAGE",
"mime_type": "image/png"
}
]
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
input=[
{"type": "text", "text": "What is in this picture?"},
{
"type": "image",
"data": "BASE64_ENCODED_IMAGE",
"mime_type": "image/png",
},
],
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
input: [
{type: 'text', text: 'What is in this picture?'},
{type: 'image', data: 'BASE64_ENCODED_IMAGE', mime_type: 'image/png'},
],
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.Content;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.ImageContent;
import com.google.genai.gaos.models.interactions.ImageContentMimeType;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.TextContent;
Client client = new Client();
List<Content> content =
List.of(
TextContent.builder().text("What is in this picture?").build(),
ImageContent.builder()
.data("BASE64_ENCODED_IMAGE")
.mimeType(ImageContentMimeType.IMAGE_PNG)
.build());
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.input(InteractionsInput.ofContent(content))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
contents := []gaos_interactions.Content{
gaos_interactions.NewContent(gaos_interactions.TextContent{
Text: "What is in this picture?",
}),
gaos_interactions.NewContent(gaos_interactions.ImageContent{
Data: genai.Ptr("BASE64_ENCODED_IMAGE"),
MimeType: gaos_interactions.ImageContentMimeTypeImagePng.ToPointer(),
}),
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput(contents)),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [
{
"type": "function",
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
}
},
"required": [
"location"
]
}
}
],
"input": "What is the weather like in Boston, MA?"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{
"type": "function",
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": (
"The city and state, e.g. San Francisco, CA"
),
}
},
"required": ["location"],
},
}],
input="What is the weather like in Boston, MA?",
)
print(response.steps[-1])
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [
{
type: 'function',
name: 'get_weather',
description: 'Get the current weather in a given location',
parameters: {
type: 'object',
properties: {
location: {
type: 'string',
description: 'The city and state, e.g. San Francisco, CA',
},
},
required: ['location'],
},
},
],
input: 'What is the weather like in Boston, MA?',
});
console.log(interaction.steps.at(-1));
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Step;
Client client = new Client();
Map<String, Object> parameters =
Map.of(
"type", "object",
"properties",
Map.of(
"location",
Map.of(
"type", "string",
"description", "The city and state, e.g. San Francisco, CA")),
"required", List.of("location"));
Function functionTool =
Function.builder()
.name("get_weather")
.description("Get the current weather in a given location")
.parameters(parameters)
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(functionTool))
.input(InteractionsInput.of("What is the weather like in Boston, MA?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
List<Step> steps = interaction.steps().orElse(List.of());
if (!steps.isEmpty()) {
System.out.println(steps.get(steps.size() - 1));
}
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
function := gaos_interactions.Function{
Name: genai.Ptr("get_weather"),
Description: genai.Ptr("Get the current weather in a given location"),
Parameters: map[string]any{
"type": "object",
"properties": map[string]any{
"location": map[string]any{
"type": "string",
"description": "The city and state, e.g. San Francisco, CA",
},
},
"required": []string{"location"},
},
}
tool := gaos_interactions.NewTool(function)
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("What is the weather like in Boston, MA?")),
Tools: []gaos_interactions.Tool{tool},
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil {
for _, step := range res.Interaction.Steps {
if step.FunctionCallStep != nil {
fmt.Println("Function Call:", step.FunctionCallStep.Name)
fmt.Println("Arguments:", step.FunctionCallStep.Arguments)
}
}
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"agent": "deep-research-preview-04-2026",
"input": "Find a cure to cancer",
"background": true
}'
from google import genai
client = genai.Client()
interaction = client.interactions.create(
agent="deep-research-preview-04-2026",
input="find a cure to cancer",
background=True,
)
print(interaction.status)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
agent: 'deep-research-preview-04-2026',
input: 'find a cure to cancer',
background: true,
});
console.log(interaction.status);
import com.google.genai.gaos.models.interactions.AgentOption;
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateAgentInteraction;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionStatus;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
CreateAgentInteraction params =
CreateAgentInteraction.builder()
.agent(AgentOption.of("deep-research-preview-04-2026"))
.input(InteractionsInput.of("find a cure to cancer"))
.background(true)
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.status().map(InteractionStatus::value).orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption("deep-research-preview-04-2026"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("find a cure to cancer")),
Background: genai.Ptr(true),
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil {
fmt.Println(res.Interaction.Status)
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"agent": "antigravity-preview-09-2026",
"input": "Read Hacker News, summarize the top 5 stories, and save results as a markdown file.",
"environment": "remote"
}'
from google import genai
client = genai.Client(http_options={"api_version": "v1beta"})
interaction = client.interactions.create(
agent="antigravity-preview-09-2026",
input=(
"Read Hacker News, summarize the top 5 stories, and save results as"
" a markdown file."
),
environment="remote",
)
print(interaction.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({apiVersion: 'v1beta'});
const interaction = await ai.interactions.create({
agent: 'antigravity-preview-09-2026',
input:
'Read Hacker News, summarize the top 5 stories, and save results as a markdown file.',
environment: 'remote',
});
console.log(interaction.output_text);
import com.google.genai.gaos.models.interactions.AgentOption;
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateAgentInteraction;
import com.google.genai.gaos.models.interactions.CreateAgentInteractionEnvironment;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.types.HttpOptions;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client =
Client.builder().httpOptions(HttpOptions.builder().apiVersion("v1beta").build()).build();
CreateAgentInteraction params =
CreateAgentInteraction.builder()
.agent(AgentOption.of("antigravity-preview-09-2026"))
.input(
InteractionsInput.of(
"Read Hacker News, summarize the top 5 stories, and save results as a markdown"
+ " file."))
.environment(CreateAgentInteractionEnvironment.of("remote"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, &genai.ClientConfig{
HTTPOptions: genai.HTTPOptions{APIVersion: "v1beta"},
})
if err != nil {
log.Fatal(err)
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption("antigravity-preview-09-2026"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Read Hacker News, summarize the top 5 stories, and save results as a markdown file.")),
Environment: genai.Ptr(gaos_interactions.NewCreateAgentInteractionEnvironment("remote")),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
# Step 1: Create an interaction with a fresh remote environment.
RESPONSE=$(curl -s -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"agent": "antigravity-preview-09-2026",
"input": "Write a hello world script at /workspace/hello.py.",
"environment": "remote"
}')
INTERACTION_ID=$(echo "$RESPONSE" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
ENV_ID=$(echo "$RESPONSE" | python3 -c "import sys,json; print(json.load(sys.stdin)['environment_id'])")
# Step 2: Reuse the same environment in a follow-up interaction.
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d "{
\"agent\": \"antigravity-preview-09-2026\",
\"input\": \"Modify the script to accept a name argument and greet the user.\",
\"environment\": \"$ENV_ID\",
\"previous_interaction_id\": \"$INTERACTION_ID\"
}"
from google import genai
client = genai.Client(http_options={"api_version": "v1beta"})
# Step 1: Create an interaction with a fresh remote environment.
interaction = client.interactions.create(
agent="antigravity-preview-09-2026",
input="Write a hello world script at /workspace/hello.py.",
environment="remote",
)
print(f"Environment ID: {interaction.environment_id}")
# Step 2: Reuse the same environment in a follow-up interaction.
interaction_2 = client.interactions.create(
agent="antigravity-preview-09-2026",
input="Modify the script to accept a name argument and greet the user.",
environment=interaction.environment_id,
previous_interaction_id=interaction.id,
)
print(interaction_2.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({apiVersion: 'v1beta'});
// Step 1: Create an interaction with a fresh remote environment.
const interaction = await ai.interactions.create({
agent: 'antigravity-preview-09-2026',
input: 'Write a hello world script at /workspace/hello.py.',
environment: 'remote',
});
console.log(`Environment ID: ${interaction.environment_id}`);
// Step 2: Reuse the same environment in a follow-up interaction.
const interaction2 = await ai.interactions.create({
agent: 'antigravity-preview-09-2026',
input: 'Modify the script to accept a name argument and greet the user.',
environment: interaction.environment_id,
previous_interaction_id: interaction.id,
});
console.log(interaction2.output_text);
import com.google.genai.gaos.models.interactions.AgentOption;
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateAgentInteraction;
import com.google.genai.gaos.models.interactions.CreateAgentInteractionEnvironment;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.environments.Environment;
import com.google.genai.types.HttpOptions;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Step;
Client client =
Client.builder().httpOptions(HttpOptions.builder().apiVersion("v1beta").build()).build();
// Step 1: Create an interaction with a fresh remote environment.
CreateAgentInteraction params1 =
CreateAgentInteraction.builder()
.agent(AgentOption.of("antigravity-preview-09-2026"))
.input(InteractionsInput.of("Write a hello world script at /workspace/hello.py."))
.environment(CreateAgentInteractionEnvironment.of("remote"))
.build();
CreateInteractionResponse response1 =
client.interactions.create(CreateInteractionRequestBody.of(params1));
Interaction interaction1 =
response1.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println("Environment ID: " + interaction1.environmentId().orElse(""));
// Step 2: Reuse the same environment in a follow-up interaction.
CreateAgentInteraction params2 =
CreateAgentInteraction.builder()
.agent(AgentOption.of("antigravity-preview-09-2026"))
.input(
InteractionsInput.of(
"Modify the script to accept a name argument and greet the user."))
.environment(
CreateAgentInteractionEnvironment.of(interaction1.environmentId().orElse("")))
.previousInteractionId(interaction1.id().orElse(null))
.build();
CreateInteractionResponse response2 =
client.interactions.create(CreateInteractionRequestBody.of(params2));
Interaction interaction2 =
response2.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction2.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, &genai.ClientConfig{
HTTPOptions: genai.HTTPOptions{APIVersion: "v1beta"},
})
if err != nil {
log.Fatal(err)
}
// Step 1: Create an interaction with a fresh remote environment.
body1 := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption("antigravity-preview-09-2026"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Write a hello world script at /workspace/hello.py.")),
Environment: genai.Ptr(gaos_interactions.NewCreateAgentInteractionEnvironment("remote")),
})
res1, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body1})
if err != nil {
log.Fatal(err)
}
if res1.Interaction == nil || res1.Interaction.EnvironmentID == nil {
log.Fatal("No environment ID returned")
}
fmt.Printf("Environment ID: %s\n", *res1.Interaction.EnvironmentID)
// Step 2: Reuse the same environment in a follow-up interaction.
body2 := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption("antigravity-preview-09-2026"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Modify the script to accept a name argument and greet the user.")),
Environment: genai.Ptr(gaos_interactions.NewCreateAgentInteractionEnvironment(*res1.Interaction.EnvironmentID)),
PreviousInteractionID: res1.Interaction.ID,
})
res2, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body2})
if err != nil {
log.Fatal(err)
}
if res2.Interaction != nil && res2.Interaction.OutputText != nil {
fmt.Println(*res2.Interaction.OutputText)
}
}
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"agent": "antigravity-preview-09-2026",
"input": "List all files under /workspace and summarize what you find.",
"environment": {
"type": "remote",
"sources": [
{
"type": "repository",
"source": "https://github.com/octocat/Spoon-Knife",
"target": "/workspace/repo"
},
{
"type": "inline",
"content": "Focus on Python files only.",
"target": "/workspace/notes.txt"
}
]
}
}'
from google import genai
client = genai.Client(http_options={"api_version": "v1beta"})
interaction = client.interactions.create(
agent="antigravity-preview-09-2026",
input="List all files under /workspace and summarize what you find.",
environment={
"type": "remote",
"sources": [
{
"type": "repository",
"source": "https://github.com/octocat/Spoon-Knife",
"target": "/workspace/repo",
},
{
"type": "inline",
"content": "Focus on Python files only.",
"target": "/workspace/notes.txt",
},
],
},
)
print(interaction.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({apiVersion: 'v1beta'});
const interaction = await ai.interactions.create({
agent: 'antigravity-preview-09-2026',
input: 'List all files under /workspace and summarize what you find.',
environment: {
type: 'remote',
sources: [
{
type: 'repository',
source: 'https://github.com/octocat/Spoon-Knife',
target: '/workspace/repo',
},
{
type: 'inline',
content: 'Focus on Python files only.',
target: '/workspace/notes.txt',
},
],
},
});
console.log(interaction.output_text);
import com.google.genai.gaos.models.interactions.AgentOption;
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateAgentInteraction;
import com.google.genai.gaos.models.interactions.CreateAgentInteractionEnvironment;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.environments.Environment;
import com.google.genai.types.HttpOptions;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Source;
import com.google.genai.gaos.models.interactions.SourceType;
Client client =
Client.builder().httpOptions(HttpOptions.builder().apiVersion("v1beta").build()).build();
com.google.genai.gaos.models.interactions.Environment env =
com.google.genai.gaos.models.interactions.Environment.builder()
.sources(
List.of(
Source.builder()
.type(SourceType.REPOSITORY)
.source("https://github.com/octocat/Spoon-Knife")
.target("/workspace/repo")
.build(),
Source.builder()
.type(SourceType.INLINE)
.content("Focus on Python files only.")
.target("/workspace/notes.txt")
.build()))
.build();
CreateAgentInteraction params =
CreateAgentInteraction.builder()
.agent(AgentOption.of("antigravity-preview-09-2026"))
.input(
InteractionsInput.of(
"List all files under /workspace and summarize what you find."))
.environment(CreateAgentInteractionEnvironment.of(env))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, &genai.ClientConfig{
HTTPOptions: genai.HTTPOptions{APIVersion: "v1beta"},
})
if err != nil {
log.Fatal(err)
}
env := gaos_interactions.Environment{
Sources: []gaos_interactions.Source{
{
Type: genai.Ptr(gaos_interactions.SourceTypeRepository),
Source: genai.Ptr("https://github.com/octocat/Spoon-Knife"),
Target: genai.Ptr("/workspace/repo"),
},
{
Type: genai.Ptr(gaos_interactions.SourceTypeInline),
Content: genai.Ptr("Focus on Python files only."),
Target: genai.Ptr("/workspace/notes.txt"),
},
},
}
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption("antigravity-preview-09-2026"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("List all files under /workspace and summarize what you find.")),
Environment: genai.Ptr(gaos_interactions.NewCreateAgentInteractionEnvironment(env)),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
# Step 1: Create a custom agent.
curl -X POST https://generativelanguage.googleapis.com/v1beta/agents \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"id": "code-reviewer",
"base_agent": "antigravity-preview-09-2026",
"system_instruction": "You are a senior code reviewer. Check every file for bugs, style issues, and security vulnerabilities.",
"base_environment": {
"type": "remote",
"sources": [{
"type": "repository",
"source": "https://github.com/octocat/Spoon-Knife",
"target": "/workspace/repo"
}]
}
}'
# Step 2: Use the custom agent.
curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"agent": "code-reviewer",
"input": "Review the latest changes in /workspace/repo/src and file a summary.",
"environment": "remote"
}'
import uuid
from google import genai
client = genai.Client(http_options={"api_version": "v1beta"})
# Step 1: Create a custom agent.
agent_id = f"code-reviewer-{uuid.uuid4().hex[:8]}"
client.agents.create(
id=agent_id,
base_agent="antigravity-preview-09-2026",
system_instruction=(
"You are a senior code reviewer. Check every file for bugs, style"
" issues, and security vulnerabilities."
),
base_environment={
"type": "remote",
"sources": [{
"type": "repository",
"source": "https://github.com/octocat/Spoon-Knife",
"target": "/workspace/repo",
}],
},
)
# Step 2: Use the custom agent.
result = client.interactions.create(
agent=agent_id,
input=(
"Review the latest changes in /workspace/repo/src and file a"
" summary."
),
environment="remote",
)
print(result.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({apiVersion: 'v1beta'});
// Step 1: Create a custom agent.
const agentId = `code-reviewer-${crypto.randomUUID().slice(0, 8)}`;
await ai.agents.create({
id: agentId,
base_agent: 'antigravity-preview-09-2026',
system_instruction:
'You are a senior code reviewer. Check every file for bugs, style issues, and security vulnerabilities.',
base_environment: {
type: 'remote',
sources: [
{
type: 'repository',
source: 'https://github.com/octocat/Spoon-Knife',
target: '/workspace/repo',
},
],
},
});
// Step 2: Use the custom agent.
const result = await ai.interactions.create({
agent: agentId,
input:
'Review the latest changes in /workspace/repo/src and file a summary.',
environment: 'remote',
});
console.log(result.output_text);
import com.google.genai.gaos.models.agents.Agent;
import com.google.genai.gaos.models.interactions.AgentOption;
import com.google.genai.gaos.models.agents.BaseEnvironment;
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateAgentInteraction;
import com.google.genai.gaos.models.interactions.CreateAgentInteractionEnvironment;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.environments.Environment;
import com.google.genai.types.HttpOptions;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Source;
import com.google.genai.gaos.models.interactions.SourceType;
import com.google.genai.gaos.models.interactions.Step;
Client client =
Client.builder().httpOptions(HttpOptions.builder().apiVersion("v1beta").build()).build();
// Step 1: Create a custom agent.
String agentId = "code-reviewer-" + UUID.randomUUID().toString().substring(0, 8);
com.google.genai.gaos.models.interactions.Environment baseEnv =
com.google.genai.gaos.models.interactions.Environment.builder()
.sources(
List.of(
Source.builder()
.type(SourceType.REPOSITORY)
.source("https://github.com/octocat/Spoon-Knife")
.target("/workspace/repo")
.build()))
.build();
Agent customAgent =
Agent.builder()
.id(agentId)
.baseAgent("antigravity-preview-09-2026")
.systemInstruction(
"You are a senior code reviewer. Check every file for bugs, style issues, and"
+ " security vulnerabilities.")
.baseEnvironment(BaseEnvironment.of(baseEnv))
.build();
client.agents.create(customAgent);
// Step 2: Use the custom agent.
CreateAgentInteraction params =
CreateAgentInteraction.builder()
.agent(AgentOption.of(agentId))
.input(
InteractionsInput.of(
"Review the latest changes in /workspace/repo/src and file a summary."))
.environment(CreateAgentInteractionEnvironment.of("remote"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"time"
"google.golang.org/genai"
gaos_agents "google.golang.org/genai/interactions/models/agents"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, &genai.ClientConfig{
HTTPOptions: genai.HTTPOptions{APIVersion: "v1beta"},
})
if err != nil {
log.Fatal(err)
}
// Step 1: Create a custom agent.
agentID := fmt.Sprintf("code-reviewer-%d", time.Now().UnixNano())
baseEnv := gaos_interactions.Environment{
Sources: []gaos_interactions.Source{
{
Type: genai.Ptr(gaos_interactions.SourceTypeRepository),
Source: genai.Ptr("https://github.com/octocat/Spoon-Knife"),
Target: genai.Ptr("/workspace/repo"),
},
},
}
_, err = client.Agents.Create(ctx, operations.CreateAgentRequest{
Body: gaos_agents.Agent{
ID: genai.Ptr(agentID),
BaseAgent: genai.Ptr("antigravity-preview-09-2026"),
SystemInstruction: genai.Ptr("You are a senior code reviewer. Check every file for bugs, style issues, and security vulnerabilities."),
BaseEnvironment: genai.Ptr(gaos_agents.NewBaseEnvironment(baseEnv)),
},
})
if err != nil {
log.Fatal(err)
}
// Step 2: Use the custom agent.
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateAgentInteraction{
Agent: gaos_interactions.AgentOption(agentID),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Review the latest changes in /workspace/repo/src and file a summary.")),
Environment: genai.Ptr(gaos_interactions.NewCreateAgentInteractionEnvironment("remote")),
})
res, err := client.Interactions.Create(ctx, operations.CreateInteractionRequest{Body: body})
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
回應¶
傳回 Interaction 資源。
範例回應
{
"created": "2025-11-26T12:25:15Z",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"model": "gemini-3.8-flash",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "Hello! I'm functioning perfectly and ready to assist you.\n\nHow are you doing today?"
}
]
}
],
"updated": "2025-11-26T12:25:15Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 7
}
],
"total_cached_tokens": 0,
"total_input_tokens": 7,
"total_output_tokens": 20,
"total_thought_tokens": 22,
"total_tokens": 49,
"total_tool_use_tokens": 0
}
}
{
"created": "2025-11-26T12:22:47Z",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"model": "gemini-3.8-flash",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "The capital of France is Paris."
}
]
}
],
"updated": "2025-11-26T12:22:47Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 50
}
],
"total_cached_tokens": 0,
"total_input_tokens": 50,
"total_output_tokens": 10,
"total_thought_tokens": 0,
"total_tokens": 60,
"total_tool_use_tokens": 0
}
}
{
"created": "2025-11-26T12:22:47Z",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"model": "gemini-3.8-flash",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "A white humanoid robot with glowing blue eyes stands holding a red skateboard."
}
]
}
],
"updated": "2025-11-26T12:22:47Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 10
},
{
"modality": "image",
"tokens": 258
}
],
"total_cached_tokens": 0,
"total_input_tokens": 268,
"total_output_tokens": 20,
"total_thought_tokens": 0,
"total_tokens": 288,
"total_tool_use_tokens": 0
}
}
{
"created": "2025-11-26T12:22:47Z",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"model": "gemini-3.8-flash",
"object": "interaction",
"status": "requires_action",
"steps": [
{
"name": "get_weather",
"type": "function_call",
"arguments": {
"location": "Boston, MA"
},
"id": "gth23981"
}
],
"updated": "2025-11-26T12:22:47Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 100
}
],
"total_cached_tokens": 0,
"total_input_tokens": 100,
"total_output_tokens": 25,
"total_thought_tokens": 0,
"total_tokens": 125,
"total_tool_use_tokens": 50
}
}
{
"agent": "deep-research-preview-04-2026",
"created": "2025-11-26T12:22:47Z",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "Here is a comprehensive research report on the current state of cancer research..."
}
]
}
],
"updated": "2025-11-26T12:22:47Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 20
}
],
"total_cached_tokens": 0,
"total_input_tokens": 20,
"total_output_tokens": 1000,
"total_thought_tokens": 500,
"total_tokens": 1520,
"total_tool_use_tokens": 0
}
}
{
"agent": "antigravity-preview-09-2026",
"created": "2025-11-26T12:22:47Z",
"environment_id": "env_abc123",
"id": "v1_ChdPU0F4YWFtNkFwS2kxZThQZ05lbXdROBIXT1NBeGFhbTZBcEtpMWU4UGdOZW13UTg",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "I've summarized the top 5 Hacker News stories and saved the results to /workspace/summary.md."
}
]
}
],
"updated": "2025-11-26T12:22:47Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 50
}
],
"total_cached_tokens": 0,
"total_input_tokens": 50,
"total_output_tokens": 500,
"total_thought_tokens": 200,
"total_tokens": 750,
"total_tool_use_tokens": 0
}
}
{
"agent": "antigravity-preview-09-2026",
"created": "2025-11-26T12:23:00Z",
"environment_id": "env_abc123",
"id": "v1_Chd2ZTJhYmNkZWZnaGlqa2xtbm9wcXJzdHV2d3h5ejAxMjM0NTY3ODkwMTIzNDU2Nzg",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "I've updated /workspace/hello.py to accept a name argument and greet the user."
}
]
}
],
"updated": "2025-11-26T12:23:00Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 80
}
],
"total_cached_tokens": 0,
"total_input_tokens": 80,
"total_output_tokens": 200,
"total_thought_tokens": 100,
"total_tokens": 380,
"total_tool_use_tokens": 0
}
}
資源¶
Interaction¶
Interaction 資源。
屬性:
id:必填 string
僅供輸出。互動完成的唯一識別碼。
**status**:必填 `string enum` 僅供輸出。互動的狀態。
可能的值: - `in_progress`:互動正在進行中。 - `requires_action`:互動需要使用者採取動作/提供輸入。 - `completed`:互動已完成。 - `failed`:互動失敗。 - `cancelled`:互動已取消。 - `incomplete`:互動已完成,但包含不完整的結果(例如達到 max_tokens)。 - `requires_continuation`:互動至少還需要一次請求才能完成。下一個請求應將 continuation_token 設為此請求傳回的權杖。created:必填 string
僅供輸出。回應的建立時間,採 ISO 8601 格式(YYYY-MM-DDThh:mm:ssZ)。
updated:必填 string
僅供輸出。回應的最後更新時間,採 ISO 8601 格式(YYYY-MM-DDThh:mm:ssZ)。
steps:必填 Step array
僅供輸出。組成互動的步驟(包含於回應中時)。
**model**:`ModelOption` 用於產生互動的 `Model` 名稱。
可能的值: - `gemma-4-26b-a4b-it`:Gemma 4 26B A4B IT - `gemma-4-31b-it`:Gemma 4 31B IT - `gemini-flash-latest`:Gemini Flash 的最新版本 - `gemini-flash-lite-latest`:Gemini Flash-Lite 的最新版本 - `gemini-pro-latest`:Gemini Pro 的最新版本 - `gemini-2.5-flash`:我們的第一個混合推理模型,支援 100 萬個權杖的上下文視窗,並具備思考預算。 - `gemini-2.5-pro`:我們最先進的多用途模型,擅長程式設計與複雜推理任務。 - `gemini-2.5-flash-lite`:我們最小、最具成本效益的模型,為大規模使用而打造。 - `gemini-3.5-flash-lite`:Gemini 3.5 Flash-Lite——我們最快速、最具成本效益的 3.5 模型,適合高輸送量執行。 - `gemini-2.5-flash-image`:我們的原生圖片生成模型,針對速度、彈性與上下文理解最佳化。文字輸入與輸出的計價與 2.5 Flash 相同。 - `gemini-3-flash-preview`:我們為速度打造的最聰明模型,結合前沿智慧與卓越的搜尋及接地能力。 - `gemini-3.1-pro-preview`:我們最新的 SOTA 推理模型,具備前所未有的深度與細膩度,以及強大的多模態理解與程式設計能力。 - `gemini-3.1-pro-preview-customtools`:Gemini 3.1 Pro Preview,針對自訂工具使用最佳化 - `gemini-3.1-flash-lite`:我們最具成本效益的模型,針對大量代理任務、翻譯與簡單資料處理最佳化。 - `gemini-3.1-flash-lite-image`:Gemini 3.1 Flash Lite Image。 - `gemini-3-pro-image`:Gemini 3 Pro Image - `nano-banana-pro-preview`:Gemini 3 Pro Image Preview - `gemini-3.1-flash-image`:Gemini 3.1 Flash Image。 - `gemini-nano-banana-2.1`:Gemini Nano Banana 2.1。 - `gemini-3.1-flash-tts-preview`:Gemini 3.1 Flash TTS:強大、低延遲的語音生成。享有自然的輸出、可引導的提示詞,以及全新的富表現力音訊標籤,精確掌控旁白。 - `gemini-3.5-flash`:Gemini 3.5 Flash——我們較早的 Flash 模型,為日常、高輸送量工作負載的速度與基礎效能而打造。 - `gemini-3.6-flash`:Gemini 3.6 Flash——我們前一代的 Flash 模型,在一般代理與日常任務中兼顧速度與多模態能力。 - `gemini-3.7-flash`:Gemini 3.7 Flash——我們高速、高效率的 Flash 模型,為日常程式設計、代理工具使用與可靠的多步驟執行而打造。 - `gemini-3.8-flash`:Gemini 3.8 Flash——我們最聰明的 Flash 模型,為長程軟體工程、自主代理與複雜的企業工作流程而設計。 - `gemini-4-argon`:Gemini 4 Argon——我們全新的前沿模型,為以深度與精確度解決最困難的問題而打造,涵蓋真實世界的軟體工程與知識工作。 - `gemini-4-carbon`:Gemini 4 Carbon——我們能力最強的推理模型,為持續的代理程式設計、多模態知識工作、科學發現與網路安全防禦而打造。 - `gemini-4-hydrogen`:Gemini 4 Hydrogen——具備前沿級智慧的模型,為程式設計、代理工作流程與多模態知識工作的快速、可靠執行而打造。 - `gemini-3.8-flash-tts`:Gemini 3.8 Flash TTS——旗艦 TTS 模型,支援 Voice Design 與雙講者劇本控制。可提示自訂聲音角色、逐行指導語音呈現,並加入聲音爆發效果。 - `gemini-3.8-flash-lite-tts`:Gemini 3.8 Flash Lite TTS——高速且具成本效益,適合快速配音、媒體在地化與高輸送量語音代理。可直接取代 gemini-3.1-flash-tts-preview。 - `gemini-3.5-transcribe`:Gemini 3.5 Transcribe——低延遲的語音轉文字模型,具備以語句為單位的語言偵測、講者分離、單字級時間戳記與自訂詞彙偏置。 - `lyria-3-clip-preview`:我們的低延遲音樂生成模型,針對高傳真音訊片段與精確的節奏控制最佳化。 - `lyria-3-pro-preview`:我們先進的完整歌曲生成模型,具備深厚的作曲理解力,針對精確的結構控制與多樣音樂風格間的複雜轉場最佳化。 - `gemini-robotics-er-1.6-preview`:Gemini Robotics-ER 1.6 Preview - `gemini-robotics-er-2-preview`:Gemini Robotics Embodied Reasoning 2 Preview - `lyria-3.5`:我們的旗艦音樂生成模型,針對具複雜結構連貫性的完整長度歌曲最佳化。 - `gemini-omni-1.2-flash`:Gemini Omni 1.2 Flash——我們最新的多模態模型,支援快速、對話式的影片生成、編輯與電影級控制。 - `gemini-omni-1.1-flash`:Gemini Omni 1.1 Flash——我們的高效能多模態模型,專為快速、對話式的影片生成、編輯與電影級控制而設計。 - `gemini-omni-flash-preview`:我們多模態模型的預覽版,支援對話式影片生成、編輯與電影級控制。**generation_config**:`GenerationConfig` 僅供輸入。模型互動的設定參數。
**seed**:`integer` 解碼時使用的種子,以確保可重現性。 **stop_sequences**:`string array` 會停止輸出互動的字元序列清單。**tool_choice**:[`ToolChoiceConfig`](/api-reference/interactions-api/create-interaction/#Resource:ToolChoiceConfig) 或 `string enum` 工具選擇設定。
**可能的類型**:**[ToolChoiceConfig](/api-reference/interactions-api/create-interaction/#Resource:ToolChoiceConfig)**
包含允許工具的工具選擇設定。**allowed_tools**:`AllowedTools` 允許的工具。
**mode**:`string enum` 工具選擇的模式。
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。**string enum**
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。**thinking_level**:`ThinkingLevel` 模型應產生的思考權杖層級。
可能的值: - `minimal`:幾乎不進行思考。 - `low`:低思考層級。 - `medium`:中思考層級。 - `high`:高思考層級。 - `xhigh`:超高思考層級,高於 `high`。**thinking_summaries**:`ThinkingSummaries` 是否在回應中包含思考摘要。
可能的值: - `auto`:自動思考摘要。 - `none`:不含思考摘要。**video_config**:`VideoConfig` 影片生成設定。
**task**:`string enum` 影片生成的選用任務模式。若未指定,模型會根據提供的文字提示詞與輸入媒體自動決定適當的模式。
可能的值: - `text_to_video`:僅從文字提示詞生成影片。 - `image_to_video`:從一或兩張來源圖片生成影片。第一張圖片定義起始影格,選用的第二張圖片定義結束影格。 - `reference_to_video`:使用參考媒體(例如圖片、音訊或影片)生成影片。 - `edit`:修改既有的輸入影片。 - `extend`:延伸既有的輸入影片。**transcription_config**:`TranscriptionConfig` 選用。語音辨識(轉錄)設定。若提供此設定,將啟用 ASR。
**language_codes**:`string array` 選用。BCP-47 語言代碼,提供音訊中所含語言的提示。若省略或留空,預設為自動語言偵測。 **custom_vocabulary**:`string array` 選用。自訂詞彙片語清單,用於引導語音辨識模型偏向辨識特定詞彙。**mode**:`TranscriptionMode` 或 `string enum` 具判別性的轉錄模式選項或列舉。
**可能的類型**:**SmartTranscriptionMode**
智慧轉錄模式的設定。 **type**:必填 `object` 值:`smart` **source_text**:`string` 要透過語音指令編輯或改寫的來源文字。提供時,模型會執行語音驅動的編輯,並輸出最終定稿的替代文字。**VerbatimTranscriptionMode**
逐字轉錄模式的設定。 **type**:必填 `object` 值:`verbatim` **timestamp_granularities**:`string array` 選用。轉錄輸出中要包含的時間戳記粒度。支援的值:"word"。若留空,則不產生時間戳記。 **diarization_mode**:`string` 選用。設定講者分離。支援的值:"speaker"。**string enum**
可能的值: - `verbatim`:逐字轉錄模式。 - `smart`:智慧轉錄模式。**structured_speech_config**:`SpeakerConfig` 或 `SpeechConfig array` 語音與多講者設定。
**可能的類型**:**SpeakerConfig**
多講者與語音生成的設定。**speakers**:`SpeechConfig array` 個別講者的設定。
**voice**:`string` 講者的聲音。 **language**:`string` 語音的語言。 **speaker**:`string` 講者的名稱,應與提示詞中給出的講者名稱相符。**SpeechConfig array**
語音互動的設定。 **voice**:`string` 講者的聲音。 **language**:`string` 語音的語言。 **speaker**:`string` 講者的名稱,應與提示詞中給出的講者名稱相符。system_instruction:string
互動的系統指令。
tools:Tool array
模型在互動期間可呼叫的工具宣告清單。
input:Content 或 Content array 或 Step array 或 string
互動的輸入。
**usage**:`Usage` 僅供輸出。互動請求的權杖用量統計。
**total_input_tokens**:`integer` 提示詞(上下文)中的權杖數量。 **initial_input_tokens**:`integer` 此次互動第一輪使用的輸入權杖數量。**input_tokens_by_modality**:`ModalityTokens array` 依模態細分的輸入權杖用量。
**modality**:`ResponseModality` 與權杖數量相關聯的模態。
可能的值: - `text`:表示模型應傳回文字。 - `image`:表示模型應傳回圖片。 - `audio`:表示模型應傳回音訊。 - `video`:表示模型應傳回影片。 - `document`:表示模型應傳回文件。**cached_tokens_by_modality**:`ModalityTokens array` 依模態細分的快取權杖用量。
**modality**:`ResponseModality` 與權杖數量相關聯的模態。
可能的值: - `text`:表示模型應傳回文字。 - `image`:表示模型應傳回圖片。 - `audio`:表示模型應傳回音訊。 - `video`:表示模型應傳回影片。 - `document`:表示模型應傳回文件。**output_tokens_by_modality**:`ModalityTokens array` 依模態細分的輸出權杖用量。
**modality**:`ResponseModality` 與權杖數量相關聯的模態。
可能的值: - `text`:表示模型應傳回文字。 - `image`:表示模型應傳回圖片。 - `audio`:表示模型應傳回音訊。 - `video`:表示模型應傳回影片。 - `document`:表示模型應傳回文件。**tool_use_tokens_by_modality**:`ModalityTokens array` 依模態細分的工具使用權杖用量。
**modality**:`ResponseModality` 與權杖數量相關聯的模態。
可能的值: - `text`:表示模型應傳回文字。 - `image`:表示模型應傳回圖片。 - `audio`:表示模型應傳回音訊。 - `video`:表示模型應傳回影片。 - `document`:表示模型應傳回文件。**grounding_tool_count**:`GroundingToolCount array` 接地工具計數。
**type**:`string enum` 與計數相關聯的接地工具類型。
可能的值: - `google_search`:使用 Google 網頁搜尋與圖片搜尋接地,以及 Web Grounding for Enterprise。 - `google_maps`:使用 Google 地圖接地。**response_format**:[`ResponseFormat`](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) 或 [`ResponseFormat`](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) `array` 強制產生的回應為符合此欄位所指定 JSON 架構的 JSON 物件。
**可能的類型**:**[ResponseFormat](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat)**
**[ResponseFormat](/api-reference/interactions-api/create-interaction/#Resource:ResponseFormat) array**
每個項目為下列其中之一的清單:`AudioResponseFormat`、`ImageResponseFormat`、`TextResponseFormat`、`VideoResponseFormat`。previous_interaction_id:string
前一次互動的 ID(如有)。
**environment**:[`EnvironmentConfig`](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentConfig) 或 `string` 互動的環境設定。可以是指定遠端環境來源的物件,或是參照既有環境 ID 的字串。
**可能的類型**:**[EnvironmentConfig](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentConfig)**
自訂環境的設定。 **type**:必填 `object` 值:`remote`**sources**:`Source array`
**type**:`string enum`
可能的值: - `gcs`:Cloud Storage 值區。 - `inline`:行內內容。 - `repository`:一般儲存庫。來源 URL 中的通訊協定前綴可識別提供者(例如 github://、gcs://)。**network**:[`EnvironmentNetworkEgressAllowlist`](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist) 或 `string enum` 環境的網路設定。
**可能的類型**:**[EnvironmentNetworkEgressAllowlist](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist)**
環境的網路輸出設定。**allowlist**:`EgressRule array` 或 `string enum` 允許的網域及其設定清單。設為 `"disabled"` 可封鎖所有網路輸出。
**可能的類型**:**EgressRule array**
單一網域允許清單規則,可選用標頭注入。 **domain**:必填 `string` 允許向外請求的網域。支援萬用字元(例如 '*.googleapis.com')。使用 '*' 可允許所有網域。 **transform**:`object` 要注入符合此規則之請求的標頭。鍵:標頭名稱(例如 "Authorization")。值:標頭值(例如 "Bearer your-token")。 **credential**:`string` 選用。以 ID 參照伺服器管理的 Credential 資源。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。**env**:`object` 或 `string` 要在沙箱環境中設定的環境變數。
**可能的類型**:**string**
直接的字串值。**mode**:`string enum` 此環境的運算生命週期模式。
可能的值: - `on_demand`:此環境的運算生命週期模式。 - `self_hosted`:此環境的運算生命週期模式。**string**
直接的字串值。environment_id:string
僅供輸出。互動的環境 ID。僅在請求中設定了環境設定時才會填入。
**service_tier**:`ServiceTier` 互動的服務層級。
可能的值: - `flex`:Flex 服務層級。 - `standard`:Standard 服務層級。 - `priority`:Priority 服務層級。**webhook_config**:`WebhookConfig` 選用。用於在互動完成時接收通知的 webhook 設定。
**uris**:`string array` 選用。若設定,這些 webhook URI 將用於 webhook 事件,以取代已註冊的 webhook。 **user_metadata**:`object` 選用。每次向 webhook 發送事件時會一併傳回的使用者中繼資料。**safety_settings**:`SafetySetting array` 互動的安全設定。
**type**:必填 `HarmCategory` 要封鎖的危害類別類型。
可能的值: - `hate_speech`:基於特定屬性宣揚暴力或煽動對個人或群體仇恨的內容。 - `dangerous_content`:宣揚、助長或促成危險活動的內容。 - `harassment`:辱罵、威脅,或意圖霸凌、折磨或嘲弄的內容。 - `sexually_explicit`:含有露骨性內容的內容。 - `jailbreak`:旨在繞過安全篩選器的提示詞。**threshold**:必填 `string enum` 封鎖內容的門檻。若危害機率超過此門檻,內容將被封鎖。
可能的值: - `block_low_and_above`:封鎖危害機率為低或更高的內容。 - `block_medium_and_above`:封鎖危害機率為中或更高的內容。 - `block_only_high`:僅封鎖危害機率為高的內容。 - `off`:完全關閉安全篩選器。**method**:`string enum` 選用。封鎖內容的方法。若未指定,預設行為是使用機率分數。
可能的值: - `severity`:危害封鎖方法同時使用機率與嚴重程度分數。 - `probability`:危害封鎖方法使用機率分數。labels:object
帶有使用者自訂中繼資料的請求標籤。標籤鍵與值的長度不得超過 63 個字元(Unicode 碼位),且只能包含小寫字母、數字字元、底線與連字號。允許國際字元。標籤值為選用。標籤鍵必須以字母開頭。
**errors**:`Error array` 僅供輸出。記錄於互動上的診斷錯誤/平台錯誤。
**code**:`string` 識別錯誤類型的 URI。 **message**:`string` 人類可讀的錯誤訊息。continuation_token:string
用於恢復長時間解碼的不透明權杖。輸出:當狀態為 REQUIRES_CONTINUATION 且可恢復解碼時設定。輸入:在 CreateInteraction 中原封不動地傳回最新權杖,以繼續解碼。
stream:boolean
僅供輸入。互動是否以串流方式傳輸。
store:boolean
僅供輸入。是否儲存回應與請求以供日後擷取。
background:boolean
僅供輸入。是否在背景執行模型互動。
**agent**:`AgentOption` 用於產生互動的 `Agent` 名稱。
可能的值: - `deep-research-preview-04-2026`:Gemini Deep Research Agent - `deep-research-max-preview-04-2026`:Gemini Deep Research Max Agent - `antigravity-preview-05-2026`:使用 Antigravity 受管理代理執行需要推理、檔案操作與工具使用的多步驟任務。 - `antigravity-preview-09-2026`:使用 Antigravity 受管理代理執行需要推理、檔案操作與工具使用的多步驟任務。**agent_config**:`AntigravityAgentConfig` 或 `DeepResearchAgentConfig` 或 `DynamicAgentConfig` 代理互動的設定參數。
**可能的類型**:**AntigravityAgentConfig**
Antigravity 代理執行階段的設定。提供伺服器端對代理執行環境與工具設定的控制。 **type**:必填 `object` 值:`antigravity` **model**:`string` 用於代理推理的模型。 **max_total_tokens**:`string` 代理執行的權杖總數上限。**DeepResearchAgentConfig**
Deep Research 代理的設定。 **type**:必填 `object` 值:`deep-research`**thinking_summaries**:`ThinkingSummaries` 是否在回應中包含思考摘要。
可能的值: - `auto`:自動思考摘要。 - `none`:不含思考摘要。**visualization**:`string enum` 是否在回應中包含視覺化內容。
可能的值: - `off`:不包含視覺化內容。 - `auto`:自動包含視覺化內容。**DynamicAgentConfig**
動態代理的設定。 **type**:必填 `object` 值:`dynamic`max_total_tokens:string
代理執行的權杖總數上限。
範例
{
"created": "2025-12-04T15:01:45Z",
"id": "v1_ChdXS0l4YWZXTk9xbk0xZThQczhEcmlROBIXV0tJeGFmV05PcW5NMWU4UHM4RHJpUTg",
"model": "gemini-3.8-flash",
"object": "interaction",
"status": "completed",
"steps": [
{
"type": "model_output",
"content": [
{
"type": "text",
"text": "Hello! I'm doing well, functioning as expected. Thank you for asking! How are you doing today?"
}
]
}
],
"updated": "2025-12-04T15:01:45Z",
"usage": {
"input_tokens_by_modality": [
{
"modality": "text",
"tokens": 7
}
],
"total_cached_tokens": 0,
"total_input_tokens": 7,
"total_output_tokens": 23,
"total_thought_tokens": 49,
"total_tokens": 79,
"total_tool_use_tokens": 0
}
}
資料模型¶
Content¶
回應的內容。
多型類型:
**AudioContent**
音訊內容區塊。 **type**:必填 `object` 值:`audio` --- **data**:`string` 音訊內容。 --- **uri**:`string` 音訊的 URI。 --- **channels**:`integer` 音訊聲道數量。 --- **sample_rate**:`integer` 音訊的取樣率。 ---**mime_type**:`string enum` 音訊的 MIME 類型。
可能的值: - `audio/wav`:WAV 音訊格式 - `audio/mp3`:MP3 音訊格式 - `audio/aiff`:AIFF 音訊格式 - `audio/aac`:AAC 音訊格式 - `audio/ogg`:OGG 音訊格式 - `audio/flac`:FLAC 音訊格式 - `audio/mpeg`:MPEG 音訊格式 - `audio/m4a`:M4A 音訊格式 - `audio/l16`:L16 音訊格式 - `audio/opus`:OPUS 音訊格式 - `audio/alaw`:ALAW 音訊格式 - `audio/mulaw`:MULAW 音訊格式 - `audio/webm`:WEBM 音訊格式{
"type": "audio",
"data": "BASE64_ENCODED_AUDIO",
"mime_type": "audio/wav"
}
**DocumentContent**
文件內容區塊。 **type**:必填 `object` 值:`document` --- **data**:`string` 文件內容。 --- **uri**:`string` 文件的 URI。 ---**mime_type**:`string enum` 文件的 MIME 類型。
可能的值: - `application/pdf`:PDF 文件格式 - `text/csv`:CSV 文件格式 - `application/vnd.openxmlformats-officedocument.presentationml.presentation`:PPTX 簡報格式 - `application/vnd.ms-powerpoint`:PPT 簡報格式 - `application/vnd.oasis.opendocument.presentation`:ODP 簡報格式 - `application/vnd.openxmlformats-officedocument.wordprocessingml.document`:DOCX 文件格式 - `application/msword`:DOC 文件格式 - `application/vnd.openxmlformats-officedocument.wordprocessingml.template`:DOTX 文件範本格式 - `application/vnd.oasis.opendocument.text`:ODT 文件格式 - `application/vnd.openxmlformats-officedocument.spreadsheetml.sheet`:XLSX 試算表格式 - `application/vnd.ms-excel`:XLS 試算表格式 - `application/vnd.oasis.opendocument.spreadsheet`:ODS 試算表格式{
"type": "document",
"data": "BASE64_ENCODED_DOCUMENT",
"mime_type": "application/pdf"
}
**[ImageContent](/api-reference/interactions-api/create-interaction/#Resource:ImageContent)**
圖片內容區塊。 **type**:必填 `object` 值:`image` --- **data**:`string` 圖片內容。 --- **uri**:`string` 圖片的 URI。 ---**resolution**:`MediaResolution` 媒體的解析度。
可能的值: - `low`:低解析度。 - `medium`:中解析度。 - `high`:高解析度。 - `ultra_high`:超高解析度。**mime_type**:`string enum` 圖片的 MIME 類型。
可能的值: - `image/png`:PNG 圖片格式 - `image/jpeg`:JPEG 圖片格式 - `image/webp`:WebP 圖片格式 - `image/heic`:HEIC 圖片格式 - `image/heif`:HEIF 圖片格式 - `image/gif`:GIF 圖片格式 - `image/bmp`:BMP 圖片格式 - `image/tiff`:TIFF 圖片格式{
"type": "image",
"data": "BASE64_ENCODED_IMAGE",
"mime_type": "image/png"
}
**[TextContent](/api-reference/interactions-api/create-interaction/#Resource:TextContent)**
文字內容區塊。 **text**:必填 `string` 文字內容。 --- **type**:必填 `object` 值:`text` ---**annotations**:`Annotation array` 模型產生內容的引用資訊。
**可能的類型**:**FileCitation**
檔案引用註釋。 **type**:必填 `object` 值:`file_citation` **document_uri**:`string` 檔案的 URI。 **file_name**:`string` 檔案的名稱。 **source**:`string` 文字某部分所歸屬的來源。 **custom_metadata**:`object` 使用者提供的、關於擷取上下文的中繼資料。 **page_number**:`integer` 被引用文件的頁碼(如適用)。 **media_id**:`string` 圖片引用時的媒體 ID(如適用)。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**PlaceCitation**
地點引用註釋。 **type**:必填 `object` 值:`place_citation` **place_id**:`string` 地點的 ID,格式為 `places/{place_id}`。 **name**:`string` 地點的標題。 **url**:`string` 地點的 URI 參照。**review_snippets**:`ReviewSnippet array` 用於產生 Google 地圖中特定地點特色相關回答的評論片段。
**title**:`string` 評論的標題。 **url**:`string` 對應至 Google 地圖上使用者評論的連結。 **review_id**:`string` 評論片段的 ID。**SpeechAnnotation**
文字內容的語音註釋。 **type**:必填 `object` 值:`speech_metadata` **speaker**:`string` 要與此輪次關聯的講者。 **style**:`string` 語音合成的風格指令。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**UrlCitation**
URL 引用註釋。 **type**:必填 `object` 值:`url_citation` **url**:`string` URL。 **title**:`string` URL 的標題。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**WordInfo**
轉錄輸出的單字級 ASR 註釋。包含單字文字、選用的時間資訊與選用的講者歸屬。 **type**:必填 `object` 值:`word_info` **text**:`string` 轉錄出的單字。 **start_offset**:`string` 單字相對於音訊起始的時間起始偏移。當 timestamp_granularities 包含 "word" 時出現。 **end_offset**:`string` 單字相對於音訊起始的時間結束偏移。當 timestamp_granularities 包含 "word" 時出現。 **speaker**:`string` 選用。此單字的講者標籤(例如 "spk_1"、"spk_2")。當 TranscriptionConfig 中設定了 diarization_mode 時出現。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。{
"type": "text",
"text": "Hello, how are you?"
}
**VideoContent**
影片內容區塊。 **type**:必填 `object` 值:`video` --- **data**:`string` 影片內容。 --- **uri**:`string` 影片的 URI。 ---**resolution**:`MediaResolution` 媒體的解析度。
可能的值: - `low`:低解析度。 - `medium`:中解析度。 - `high`:高解析度。 - `ultra_high`:超高解析度。**processing**:`MediaProcessing` 或 `string enum` 模型如何處理此影片以進行理解。
**可能的類型**:**MediaProcessing**
**type**:必填 `object` 值:`static` **start_offset**:`string` 選用。區段起始時間。以十進位秒數加上 's' 後綴指定,例如 "10.5s"。必須為非負數。 **end_offset**:`string` 選用。區段結束時間。以十進位秒數加上 's' 後綴指定,例如 "30s"。必須為非負數,且在設定 `start_offset` 時須大於 `start_offset`。 **fps**:`number` 選用。影片影格率取樣密度。**string enum**
可能的值: - `static`:固定速率的影格擷取。所有影格皆置於上下文中。 - `agentic`:由模型驅動的動態導覽。**mime_type**:`string enum` 影片的 MIME 類型。
可能的值: - `video/mp4`:MP4 影片格式 - `video/mpeg`:MPEG 影片格式 - `video/mpg`:MPG 影片格式 - `video/mov`:MOV 影片格式 - `video/avi`:AVI 影片格式 - `video/x-flv`:FLV 影片格式 - `video/webm`:WebM 影片格式 - `video/wmv`:WMV 影片格式 - `video/3gpp`:3GPP 影片格式{
"type": "video",
"uri": "https://www.youtube.com/watch?v=9hE5-98ZeCg"
}
Tool¶
模型可使用的工具。
多型類型:
**CodeExecution**
模型可用來執行程式碼的工具。 **type**:必填 `object` 值:`code_execution` **範例** === "REST"curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "code_execution"
}],
"input": "Calculate the first 10 Fibonacci numbers"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{"type": "code_execution"}],
input="Calculate the first 10 Fibonacci numbers",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [{type: 'code_execution'}],
input: 'Calculate the first 10 Fibonacci numbers',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CodeExecution;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(new CodeExecution()))
.input(InteractionsInput.of("Calculate the first 10 Fibonacci numbers"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.CodeExecution{})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Calculate the first 10 Fibonacci numbers")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**ComputerUse**
模型可用來與電腦互動的工具。 **type**:必填 `object` 值:`computer_use` ---**environment**:`string enum` 正在操作的環境。
可能的值: - `browser`:在網頁瀏覽器中操作。 - `mobile`:在行動環境中操作。 - `desktop`:在桌面環境中操作。**disabled_safety_policies**:`string enum array` 選用。computer use 停用的安全政策。
可能的值: - `financial_transactions`:金融交易的安全政策。 - `sensitive_data_modification`:敏感資料修改的安全政策。 - `communication_tool`:通訊工具(例如 Gmail、Chat、Meet)的安全政策。 - `account_creation`:帳號建立的安全政策。 - `data_modification`:資料修改的安全政策。 - `user_consent_management`:使用者同意管理的安全政策。 - `legal_terms_and_agreements`:法律條款與協議的安全政策。curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "computer_use"
}],
"input": "Find a flight to Tokyo"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{"type": "computer_use"}],
input="Find a flight to Tokyo",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [{type: 'computer_use'}],
input: 'Find a flight to Tokyo',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.ComputerUse;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(new ComputerUse()))
.input(InteractionsInput.of("Find a flight to Tokyo"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.ComputerUse{})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Find a flight to Tokyo")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**FileSearch**
模型可用來搜尋檔案的工具。 **type**:必填 `object` 值:`file_search` --- **file_search_store_names**:`string array` 要搜尋的 File Search 儲存庫名稱。 --- **top_k**:`integer` 要擷取的語意檢索區塊數量。 --- **metadata_filter**:`string` 要套用至語意檢索文件與區塊的中繼資料篩選條件。 **範例** === "REST"curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "file_search",
"file_search_store_names": ["fileSearchStores/m64d1sevsr4y-xfyawui3fxqg"]
}],
"input": "Who is the author of the book?"
}'
from google import genai
client = genai.Client()
# Create a file search store so we have a valid one to use.
store = client.file_search_stores.create()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[
{"type": "file_search", "file_search_store_names": [store.name]}
],
input="What documents are available?",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
// Create a file search store so we have a valid one to use.
const store = await ai.fileSearchStores.create({});
if (!store.name) {
throw new Error('Store creation failed: Name is undefined');
}
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [
{
type: 'file_search',
file_search_store_names: [store.name],
},
],
input: 'What documents are available?',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.types.CreateFileSearchStoreConfig;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.FileSearch;
import com.google.genai.types.FileSearchStore;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
// Create a file search store so we have a valid one to use.
FileSearchStore store =
client.fileSearchStores.create(CreateFileSearchStoreConfig.builder().build());
String storeName = store.name().orElseThrow();
FileSearch tool = FileSearch.builder().fileSearchStoreNames(List.of(storeName)).build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(tool))
.input(InteractionsInput.of("What documents are available?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.FileSearch{
FileSearchStoreNames: []string{store.Name},
})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("What documents are available?")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**Function**
模型可使用的工具。 **type**:必填 `object` 值:`function` --- **name**:`string` 函式的名稱。 --- **description**:`string` 函式的說明。 --- **parameters**:`object` 函式參數的 JSON Schema。支援 JSON Schema 的子集。`minLength`、`maxLength`、`minProperties`、`maxProperties`、`uniqueItems`、`contains`、`minContains` 與 `maxContains` 會被忽略且不強制執行。 **範例** === "REST"curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "function",
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
}
},
"required": ["location"]
}
}],
"input": "What is the weather like in Boston, MA?"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{
"type": "function",
"name": "get_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": (
"The city and state, e.g. San Francisco, CA"
),
}
},
"required": ["location"],
},
}],
input="What is the weather like in Boston?",
)
print(response.steps[-1])
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [
{
type: 'function',
name: 'get_weather',
description: 'Get the current weather in a given location',
parameters: {
type: 'object',
properties: {
location: {
type: 'string',
description: 'The city and state, e.g. San Francisco, CA',
},
},
required: ['location'],
},
},
],
input: 'What is the weather like in Boston?',
});
console.log(interaction.steps.at(-1));
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Step;
Client client = new Client();
Map<String, Object> parameters =
Map.of(
"type", "object",
"properties",
Map.of(
"location",
Map.of(
"type", "string",
"description", "The city and state, e.g. San Francisco, CA")),
"required", List.of("location"));
Function functionTool =
Function.builder()
.name("get_weather")
.description("Get the current weather in a given location")
.parameters(parameters)
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(functionTool))
.input(InteractionsInput.of("What is the weather like in Boston?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
List<Step> steps = interaction.steps().orElse(List.of());
if (!steps.isEmpty()) {
System.out.println(steps.get(steps.size() - 1));
}
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
function := gaos_interactions.Function{
Name: genai.Ptr("get_weather"),
Description: genai.Ptr("Get the current weather in a given location"),
Parameters: map[string]any{
"type": "object",
"properties": map[string]any{
"location": map[string]any{
"type": "string",
"description": "The city and state, e.g. San Francisco, CA",
},
},
"required": []string{"location"},
},
}
tool := gaos_interactions.NewTool(function)
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("What is the weather like in Boston, MA?")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil {
for _, step := range res.Interaction.Steps {
if step.FunctionCallStep != nil {
fmt.Println("Function Call:", step.FunctionCallStep.Name)
fmt.Println("Arguments:", step.FunctionCallStep.Arguments)
}
}
}
}
**GoogleMaps**
模型可用來呼叫 Google 地圖的工具。 **type**:必填 `object` 值:`google_maps` --- **enable_widget**:`boolean` 是否在回應的工具呼叫結果中傳回 widget 上下文權杖。 --- **latitude**:`number` 使用者位置的緯度。 --- **longitude**:`number` 使用者位置的經度。 **範例** === "REST"curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "google_maps",
"latitude": 37.7749,
"longitude": -122.4194
}],
"input": "What is the best food near me?"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[
{"type": "google_maps", "latitude": 37.7749, "longitude": -122.4194}
],
input="What is the best food near me?",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [
{
type: 'google_maps',
latitude: 37.7749,
longitude: -122.4194,
},
],
input: 'What is the best food near me?',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.GoogleMaps;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
GoogleMaps tool = GoogleMaps.builder().latitude(37.7749).longitude(-122.4194).build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(tool))
.input(InteractionsInput.of("What is the best food near me?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.GoogleMaps{
Latitude: genai.Ptr(37.7749),
Longitude: genai.Ptr(-122.4194),
})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("What is the best food near me?")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**GoogleSearch**
模型可用來搜尋 Google 的工具。 **type**:必填 `object` 值:`google_search` ---**search_types**:`string enum array` 要啟用的搜尋接地類型。
可能的值: - `web_search`:設定此欄位可啟用網頁搜尋。僅傳回文字結果。 - `image_search`:設定此欄位可啟用圖片搜尋。傳回圖片位元組。curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "google_search"
}],
"input": "Who is the current president of France?"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{"type": "google_search"}],
input="Who is the current president of France?",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [{type: 'google_search'}],
input: 'Who is the current president of France?',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.GoogleSearch;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
Client client = new Client();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(new GoogleSearch()))
.input(InteractionsInput.of("Who is the current president of France?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.GoogleSearch{})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Who is the current president of France?")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**McpServer**
MCPServer 是可由模型呼叫以執行動作的伺服器。 **type**:必填 `object` 值:`mcp_server` --- **name**:`string` MCPServer 的名稱。 --- **url**:`string` MCPServer 端點的完整 URL。範例:"https://api.example.com/mcp" --- **headers**:`object` 選用:如有需要,用於驗證標頭、逾時等欄位。 ---**allowed_tools**:`AllowedTools array` 允許的工具。
**mode**:`string enum` 工具選擇的模式。
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "mcp_server",
"name": "weather_service",
"url": "https://gemini-api-demos.uc.r.appspot.com/mcp"
}],
"input": "Today is 12-05-2025, what is the temperature today in London?"
}'
import os
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{
"type": "mcp_server",
"name": "weather_service",
"url": "https://gemini-api-demos.uc.r.appspot.com/mcp",
}],
input="Today is 12-05-2025, what is the temperature today in London?",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [
{
type: 'mcp_server',
name: 'weather_service',
url: 'https://gemini-api-demos.uc.r.appspot.com/mcp',
},
],
input: 'Today is 12-05-2025, what is the temperature today in London?',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.MCPServer;
Client client = new Client();
MCPServer mcpTool =
MCPServer.builder()
.name("weather_service")
.url("https://gemini-api-demos.uc.r.appspot.com/mcp")
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(mcpTool))
.input(
InteractionsInput.of(
"Today is 12-05-2025, what is the temperature today in London?"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
mcpServer := gaos_interactions.MCPServer{
Name: genai.Ptr("weather_service"),
URL: genai.Ptr("https://gemini-api-demos.uc.r.appspot.com/mcp"),
}
tool := gaos_interactions.NewTool(mcpServer)
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Today is 12-05-2025, what is the temperature today in London?")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
**UrlContext**
模型可用來擷取 URL 上下文的工具。 **type**:必填 `object` 值:`url_context` **範例** === "REST"curl -X POST https://generativelanguage.googleapis.com/v1beta/interactions \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.8-flash",
"tools": [{
"type": "url_context"
}],
"input": "Summarize https://www.example.com"
}'
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.8-flash",
tools=[{"type": "url_context"}],
input="Summarize https://www.example.com",
)
print(response.output_text)
import {GoogleGenAI} from '@google/genai';
const ai = new GoogleGenAI({});
const interaction = await ai.interactions.create({
model: 'gemini-3.8-flash',
tools: [{type: 'url_context'}],
input: 'Summarize https://www.example.com',
});
console.log(interaction.output_text);
import com.google.genai.Client;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import com.google.genai.gaos.models.operations.CreateInteractionResponse;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.URLContext;
Client client = new Client();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model("gemini-3.8-flash")
.tools(List.of(new URLContext()))
.input(InteractionsInput.of("Summarize https://www.example.com"))
.build();
CreateInteractionResponse response =
client.interactions.create(CreateInteractionRequestBody.of(params));
Interaction interaction =
response.interaction().orElseThrow(() -> new RuntimeException("No interaction returned"));
System.out.println(interaction.outputText().orElse(""));
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
gaos_interactions "google.golang.org/genai/interactions/models/interactions"
"google.golang.org/genai/interactions/models/operations"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
tool := gaos_interactions.NewTool(gaos_interactions.URLContext{})
body := operations.NewCreateInteractionRequestBody(gaos_interactions.CreateModelInteraction{
Model: gaos_interactions.Model("gemini-3.8-flash"),
Input: genai.Ptr(gaos_interactions.NewInteractionsInput("Summarize https://www.example.com")),
Tools: []gaos_interactions.Tool{tool},
})
req := operations.CreateInteractionRequest{
Body: body,
}
res, err := client.Interactions.Create(ctx, req)
if err != nil {
log.Fatal(err)
}
if res.Interaction != nil && res.Interaction.OutputText != nil {
fmt.Println(*res.Interaction.OutputText)
}
}
ResponseFormat¶
多型類型:
**AudioResponseFormat**
音訊輸出格式的設定。 **type**:必填 `object` 值:`audio` ---**mime_type**:`string enum` 音訊輸出的 MIME 類型。
可能的值: - `audio/mp3`:MP3 音訊格式。 - `audio/ogg_opus`:OGG Opus 音訊格式。 - `audio/l16`:原始 PCM(L16)音訊格式。 - `audio/wav`:WAV 音訊格式。 - `audio/alaw`:A-law 音訊格式。 - `audio/mulaw`:Mu-law 音訊格式。**delivery**:`string enum` 音訊輸出的傳遞模式。
可能的值: - `inline`:音訊資料以行內方式在回應中傳回。 - `uri`:音訊資料以 URI 形式傳回。{
"type": "audio",
"sample_rate": 24000
}
**ImageResponseFormat**
圖片輸出格式的設定。 **type**:必填 `object` 值:`image` ---**mime_type**:`string enum` 圖片輸出的 MIME 類型。
可能的值: - `image/jpeg`:JPEG 圖片格式。**delivery**:`string enum` 圖片輸出的傳遞模式。
可能的值: - `inline`:圖片資料以行內方式在回應中傳回。 - `uri`:圖片資料以 URI 形式傳回。**aspect_ratio**:`string enum` 圖片輸出的長寬比。
可能的值: - `1:1`:1:1 長寬比。 - `2:3`:2:3 長寬比。 - `3:2`:3:2 長寬比。 - `3:4`:3:4 長寬比。 - `4:3`:4:3 長寬比。 - `4:5`:4:5 長寬比。 - `5:4`:5:4 長寬比。 - `9:16`:9:16 長寬比。 - `16:9`:16:9 長寬比。 - `21:9`:21:9 長寬比。 - `1:8`:1:8 長寬比。 - `8:1`:8:1 長寬比。 - `1:4`:1:4 長寬比。 - `4:1`:4:1 長寬比。**image_size**:`string enum` 圖片輸出的大小。
可能的值: - `512`:512px 圖片大小。 - `1K`:1K 圖片大小。 - `2K`:2K 圖片大小。 - `4K`:4K 圖片大小。{
"type": "image",
"aspect_ratio": "16:9",
"image_size": "1K",
"mime_type": "image/jpeg"
}
**TextResponseFormat**
文字輸出格式的設定。 **type**:必填 `object` 值:`text` ---**mime_type**:`string enum` 文字輸出的 MIME 類型。
可能的值: - `application/json`:JSON 輸出格式。 - `text/plain`:純文字輸出格式。{
"type": "text",
"mime_type": "application/json",
"schema": {
"type": "object",
"properties": {
"ingredients": {
"type": "array",
"items": {
"type": "string"
}
},
"recipe_name": {
"type": "string"
}
},
"required": [
"ingredients",
"recipe_name"
]
}
}
**VideoResponseFormat**
影片輸出格式的設定。 **type**:必填 `object` 值:`video` ---**delivery**:`string enum` 影片輸出的傳遞模式。
可能的值: - `inline`:影片資料以行內方式在回應中傳回。 - `uri`:影片資料以 URI 形式傳回。**aspect_ratio**:`string enum` 影片輸出的長寬比。
可能的值: - `16:9`:16:9 長寬比。 - `9:16`:9:16 長寬比。**resolution**:`string enum` 影片輸出解析度。預設為 720p。
可能的值: - `360p`:360p 解析度。 - `720p`:720p 解析度。 - `1080p`:1080p 解析度。 - `4k`:4K 解析度。{
"type": "video",
"aspect_ratio": "16:9",
"delivery": "inline"
}
Step¶
互動中的一個步驟。
多型類型:
**CodeExecutionCallStep**
程式碼執行呼叫步驟。**arguments**:必填 `CodeExecutionCallStepArguments` 要傳遞給程式碼執行的引數。
**language**:`string enum` `code` 的程式語言。
可能的值: - `python`:Python >= 3.10,可使用 numpy 與 simpy。 - `bash`:Bash。僅適用於 Gemini 3.0 以上模型。{
"type": "code_execution_call",
"arguments": {
"code": "print(sum(range(1, 11)))"
},
"id": "code_call_71021"
}
**CodeExecutionResultStep**
程式碼執行結果步驟。**result**:必填 [`Content`](/api-reference/interactions-api/create-interaction/#Resource:Content) `array` 或 `string` 程式碼執行的輸出。
**可能的類型**:**[Content](/api-reference/interactions-api/create-interaction/#Resource:Content) array**
清單的內容。**string**
直接的字串值。{
"type": "code_execution_result",
"call_id": "code_call_71021",
"result": "55\n"
}
**FileSearchCallStep**
File Search 呼叫步驟。 **id**:必填 `string` 此特定工具呼叫的唯一 ID。 --- **type**:必填 `object` 值:`file_search_call` --- **signature**:`string` 用於後端驗證的簽章雜湊。 **範例** **FileSearchCallStep**{
"type": "file_search_call",
"id": "file_call_88192"
}
**FileSearchResultStep**
File Search 結果步驟。 **call_id**:必填 `string` 用於對應函式呼叫區塊中 ID 的 ID。 --- **type**:必填 `object` 值:`file_search_result` --- **signature**:`string` 用於後端驗證的簽章雜湊。 **範例** **FileSearchResultStep**{
"type": "file_search_result",
"call_id": "file_call_88192"
}
**FunctionCallStep**
函式工具呼叫步驟。 **name**:必填 `string` 要呼叫的工具名稱。 --- **arguments**:必填 `object` 要傳遞給函式的引數。 --- **id**:必填 `string` 此特定工具呼叫的唯一 ID。 --- **type**:必填 `object` 值:`function_call` **範例** **FunctionCallStep**{
"name": "get_weather",
"type": "function_call",
"arguments": {
"location": "Boston, MA"
},
"id": "call_98231"
}
**FunctionResultStep**
函式工具呼叫的結果。**result**:必填 `(`[`ImageContent`](/api-reference/interactions-api/create-interaction/#Resource:ImageContent) 或 [`TextContent`](/api-reference/interactions-api/create-interaction/#Resource:TextContent)`) array` 或 `object` 或 `string` 工具呼叫的結果。
**可能的類型**:**([ImageContent](/api-reference/interactions-api/create-interaction/#Resource:ImageContent) 或 [TextContent](/api-reference/interactions-api/create-interaction/#Resource:TextContent)) array**
每個項目為下列其中之一的清單:[`ImageContent`](/api-reference/interactions-api/create-interaction/#Resource:ImageContent)、[`TextContent`](/api-reference/interactions-api/create-interaction/#Resource:TextContent)。**string**
直接的字串值。{
"name": "get_weather",
"type": "function_result",
"call_id": "call_98231",
"result": [
{
"type": "text",
"text": "{\"weather\":\"sunny\"}"
}
]
}
**GoogleMapsCallStep**
Google 地圖呼叫步驟。 **id**:必填 `string` 此特定工具呼叫的唯一 ID。 --- **type**:必填 `object` 值:`google_maps_call` ---**arguments**:`GoogleMapsCallStepArguments` 要傳遞給 Google 地圖工具的引數。
**queries**:`string array` 要執行的查詢。{
"type": "google_maps_call",
"arguments": {
"latitude": 37.7749,
"longitude": -122.4194
},
"id": "maps_call_39201"
}
**GoogleMapsResultStep**
Google 地圖結果步驟。**result**:必填 `GoogleMapsResultItem array`
**places**:`GoogleMapsResultPlaces array`
**place_id**:`string` **name**:`string` **url**:`string`**review_snippets**:`ReviewSnippet array`
**title**:`string` 評論的標題。 **url**:`string` 對應至 Google 地圖上使用者評論的連結。 **review_id**:`string` 評論片段的 ID。{
"type": "google_maps_result",
"call_id": "maps_call_39201",
"result": [
{
"name": "Golden Gate Park",
"place_id": "ChIJIQBpAG2ahYAR9R7bNdTLg8M",
"rating": 4.8
}
]
}
**GoogleSearchCallStep**
Google 搜尋呼叫步驟。**arguments**:必填 `GoogleSearchCallStepArguments` 要傳遞給 Google 搜尋的引數。
**queries**:`string array` 後續網頁搜尋的網頁搜尋查詢。**search_type**:`string enum` 啟用的搜尋接地類型。
可能的值: - `web_search`:設定此欄位可啟用網頁搜尋。僅傳回文字結果。 - `image_search`:設定此欄位可啟用圖片搜尋。傳回圖片位元組。{
"type": "google_search_call",
"arguments": {
"query": "Who won the men's 100m in Paris 2024?"
},
"id": "search_call_19201"
}
**GoogleSearchResultStep**
Google 搜尋結果步驟。**result**:必填 `GoogleSearchResultItem array` Google 搜尋的結果。
**search_suggestions**:`string` 可嵌入網頁或應用程式 webview 的網頁內容片段。{
"type": "google_search_result",
"call_id": "search_call_19201",
"result": [
{
"title": "Paris 2024 Olympics: Noah Lyles wins men's 100m gold",
"url": "https://olympics.com/en/news/paris-2024-noah-lyles-wins-mens-100m-gold",
"snippet": "American Noah Lyles won the Olympic men's 100m gold medal in a photo finish."
}
]
}
**McpServerToolCallStep**
MCPServer 工具呼叫步驟。 **name**:必填 `string` 被呼叫的工具名稱。 --- **server_name**:必填 `string` 所使用的 MCP 伺服器名稱。 --- **arguments**:必填 `object` 函式引數的 JSON 物件。 --- **id**:必填 `string` 此特定工具呼叫的唯一 ID。 --- **type**:必填 `object` 值:`mcp_server_tool_call` **範例** **McpServerToolCallStep**{
"name": "calculate_tax",
"type": "mcp_server_tool_call",
"arguments": {
"income": 120000,
"state": "CA"
},
"id": "mcp_call_29012",
"server_name": "financial_mcp_server"
}
**McpServerToolResultStep**
MCPServer 工具結果步驟。**result**:必填 `(`[`ImageContent`](/api-reference/interactions-api/create-interaction/#Resource:ImageContent) 或 [`TextContent`](/api-reference/interactions-api/create-interaction/#Resource:TextContent)`) array` 或 `object` 或 `string` MCP 伺服器呼叫的輸出。可以是簡單文字或豐富內容。
**可能的類型**:**([ImageContent](/api-reference/interactions-api/create-interaction/#Resource:ImageContent) 或 [TextContent](/api-reference/interactions-api/create-interaction/#Resource:TextContent)) array**
每個項目為下列其中之一的清單:[`ImageContent`](/api-reference/interactions-api/create-interaction/#Resource:ImageContent)、[`TextContent`](/api-reference/interactions-api/create-interaction/#Resource:TextContent)。**string**
直接的字串值。{
"name": "calculate_tax",
"type": "mcp_server_tool_result",
"call_id": "mcp_call_29012",
"result": {
"tax_due": 32400
},
"server_name": "financial_mcp_server"
}
**ModelOutputStep**
模型產生的輸出。 **type**:必填 `object` 值:`model_output` --- **content**:[`Content`](/api-reference/interactions-api/create-interaction/#Resource:Content) `array` **範例** **ModelOutputStep**{
"type": "model_output",
"content": [
{
"type": "text",
"text": "The capital of France is Paris."
}
]
}
**ProcessingCallStep**
由伺服器發起、用於媒體分析(例如影片理解)的處理步驟。 **id**:必填 `string` 此特定工具呼叫的唯一 ID。 --- **type**:必填 `object` 值:`processing_call` --- **signature**:`string` 用於後端驗證的簽章雜湊。**ProcessingResultStep**
由伺服器發起之媒體處理步驟的結果。 **call_id**:必填 `string` 用於對應函式呼叫區塊中 ID 的 ID。 --- **type**:必填 `object` 值:`processing_result` --- **signature**:`string` 用於後端驗證的簽章雜湊。**ThoughtStep**
思考步驟。 **type**:必填 `object` 值:`thought` --- **signature**:`string` 用於後端驗證的簽章雜湊。 --- **summary**:[`Content`](/api-reference/interactions-api/create-interaction/#Resource:Content) `array` 思考的摘要。 **範例** **ThoughtStep**{
"type": "thought",
"signature": "thought_sig_abcd1234",
"summary": [
{
"type": "text",
"text": "The model is searching Google for the capital of France."
}
]
}
**UrlContextCallStep**
URL 上下文呼叫步驟。**arguments**:必填 `UrlContextCallArguments` 要傳遞給 URL 上下文的引數。
**urls**:`string array` 要擷取的 URL。{
"type": "url_context_call",
"arguments": {
"urls": [
"https://www.example.com"
]
},
"id": "url_call_10219"
}
**UrlContextResultStep**
URL 上下文結果步驟。**result**:必填 `UrlContextResult array` URL 上下文的結果。
**url**:`string` 已擷取的 URL。**status**:`string enum` URL 擷取的狀態。
可能的值: - `success`:URL 擷取成功。 - `error`:URL 擷取因錯誤而失敗。 - `paywall`:URL 擷取失敗,因為內容位於付費牆之後。 - `unsafe`:URL 擷取失敗,因為內容不安全。{
"type": "url_context_result",
"call_id": "url_call_10219",
"result": [
{
"title": "Example Domain",
"url": "https://www.example.com",
"snippet": "This domain is for use in illustrative examples in documents."
}
]
}
**UserInputStep**
使用者提供的輸入。 **type**:必填 `object` 值:`user_input` --- **content**:[`Content`](/api-reference/interactions-api/create-interaction/#Resource:Content) `array` **範例** **UserInputStep**{
"type": "user_input",
"content": [
{
"type": "text",
"text": "What is the capital of France?"
}
]
}
EnvironmentConfig¶
自訂環境的設定。
屬性:
type:必填 object
值:remote
**sources**:`Source array`
**type**:`string enum`
可能的值: - `gcs`:Cloud Storage 值區。 - `inline`:行內內容。 - `repository`:一般儲存庫。來源 URL 中的通訊協定前綴可識別提供者(例如 github://、gcs://)。**network**:[`EnvironmentNetworkEgressAllowlist`](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist) 或 `string enum` 環境的網路設定。
**可能的類型**:**[EnvironmentNetworkEgressAllowlist](/api-reference/interactions-api/create-interaction/#Resource:EnvironmentNetworkEgressAllowlist)**
環境的網路輸出設定。**allowlist**:`EgressRule array` 或 `string enum` 允許的網域及其設定清單。設為 `"disabled"` 可封鎖所有網路輸出。
**可能的類型**:**EgressRule array**
單一網域允許清單規則,可選用標頭注入。 **domain**:必填 `string` 允許向外請求的網域。支援萬用字元(例如 '*.googleapis.com')。使用 '*' 可允許所有網域。 **transform**:`object` 要注入符合此規則之請求的標頭。鍵:標頭名稱(例如 "Authorization")。值:標頭值(例如 "Bearer your-token")。 **credential**:`string` 選用。以 ID 參照伺服器管理的 Credential 資源。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。environment_id:string
選用。互動的環境 ID。若指定,請求將更新既有環境,而非建立新環境。
**env**:`object` 或 `string` 要在沙箱環境中設定的環境變數。
**可能的類型**:**string**
直接的字串值。**mode**:`string enum` 此環境的運算生命週期模式。
可能的值: - `on_demand`:此環境的運算生命週期模式。 - `self_hosted`:此環境的運算生命週期模式。custom_config:object
選用。任意由客戶定義的 JSON 設定,會透傳至自行託管環境的輪詢器。僅在 mode 為 SELF_HOSTED 時支援。
範例
{
"type": "remote",
"sources": [
{
"type": "inline",
"content": "You are a data analyst. Always include visualizations and export results as PDF.",
"target": ".agents/AGENTS.md"
},
{
"type": "inline",
"content": "---\nname: slide-maker\ndescription: Create HTML slide decks\n---\n# Slide Maker\n\nWhen asked to create a presentation:\n1. Analyze the input data\n2. Create an HTML slide deck with reveal.js\n3. Save to /workspace/output/slides.html",
"target": ".agents/skills/slide-maker/SKILL.md"
}
]
}
{
"type": "remote",
"sources": [
{
"type": "repository",
"source": "https://github.com/my-org/my-skills.git",
"target": ".agents/skills"
},
{
"type": "gcs",
"source": "gs://my-bucket/my-folder",
"target": "/workspace/data"
}
]
}
{
"type": "remote",
"network": {
"allowlist": [
{
"domain": "pypi.org"
},
{
"domain": "*.github.com"
}
]
}
}
{
"type": "remote",
"network": {
"allowlist": [
{
"domain": "api.github.com",
"transform": {
"Authorization": "Bearer YOUR_GITHUB_TOKEN"
}
}
]
}
}
EnvironmentNetworkEgressAllowlist¶
環境的網路輸出設定。
屬性:
**allowlist**:`EgressRule array` 或 `string enum` 允許的網域及其設定清單。設為 `"disabled"` 可封鎖所有網路輸出。
**可能的類型**:**EgressRule array**
單一網域允許清單規則,可選用標頭注入。 **domain**:必填 `string` 允許向外請求的網域。支援萬用字元(例如 '*.googleapis.com')。使用 '*' 可允許所有網域。 **transform**:`object` 要注入符合此規則之請求的標頭。鍵:標頭名稱(例如 "Authorization")。值:標頭值(例如 "Bearer your-token")。 **credential**:`string` 選用。以 ID 參照伺服器管理的 Credential 資源。**string enum**
可能的值: - `disabled`:所有網路輸出皆被封鎖。範例
網路允許清單
{
"allowlist": [
{
"domain": "github.com",
"transform": [
{
"Authorization": "Bearer your-token"
}
]
},
{
"domain": "*.googleapis.com"
}
]
}
ToolChoiceConfig¶
包含允許工具的工具選擇設定。
屬性:
**allowed_tools**:`AllowedTools` 允許的工具。
**mode**:`string enum` 工具選擇的模式。
可能的值: - `auto`:自動工具選擇。 - `any`:任意工具選擇。 - `none`:不使用工具選擇。 - `validated`:經驗證的工具選擇。範例
{
"allowed_tools": {
"mode": "any",
"tools": [
"my_tool"
]
}
}
ImageContent¶
圖片內容區塊。
屬性:
type:必填 object
值:image
data:string
圖片內容。
uri:string
圖片的 URI。
**resolution**:`MediaResolution` 媒體的解析度。
可能的值: - `low`:低解析度。 - `medium`:中解析度。 - `high`:高解析度。 - `ultra_high`:超高解析度。**mime_type**:`string enum` 圖片的 MIME 類型。
可能的值: - `image/png`:PNG 圖片格式 - `image/jpeg`:JPEG 圖片格式 - `image/webp`:WebP 圖片格式 - `image/heic`:HEIC 圖片格式 - `image/heif`:HEIF 圖片格式 - `image/gif`:GIF 圖片格式 - `image/bmp`:BMP 圖片格式 - `image/tiff`:TIFF 圖片格式範例
圖片
{
"type": "image",
"data": "BASE64_ENCODED_IMAGE",
"mime_type": "image/png"
}
TextContent¶
文字內容區塊。
屬性:
text:必填 string
文字內容。
type:必填 object
值:text
**annotations**:`Annotation array` 模型產生內容的引用資訊。
**可能的類型**:**FileCitation**
檔案引用註釋。 **type**:必填 `object` 值:`file_citation` **document_uri**:`string` 檔案的 URI。 **file_name**:`string` 檔案的名稱。 **source**:`string` 文字某部分所歸屬的來源。 **custom_metadata**:`object` 使用者提供的、關於擷取上下文的中繼資料。 **page_number**:`integer` 被引用文件的頁碼(如適用)。 **media_id**:`string` 圖片引用時的媒體 ID(如適用)。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**PlaceCitation**
地點引用註釋。 **type**:必填 `object` 值:`place_citation` **place_id**:`string` 地點的 ID,格式為 `places/{place_id}`。 **name**:`string` 地點的標題。 **url**:`string` 地點的 URI 參照。**review_snippets**:`ReviewSnippet array` 用於產生 Google 地圖中特定地點特色相關回答的評論片段。
**title**:`string` 評論的標題。 **url**:`string` 對應至 Google 地圖上使用者評論的連結。 **review_id**:`string` 評論片段的 ID。**SpeechAnnotation**
文字內容的語音註釋。 **type**:必填 `object` 值:`speech_metadata` **speaker**:`string` 要與此輪次關聯的講者。 **style**:`string` 語音合成的風格指令。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**UrlCitation**
URL 引用註釋。 **type**:必填 `object` 值:`url_citation` **url**:`string` URL。 **title**:`string` URL 的標題。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。**WordInfo**
轉錄輸出的單字級 ASR 註釋。包含單字文字、選用的時間資訊與選用的講者歸屬。 **type**:必填 `object` 值:`word_info` **text**:`string` 轉錄出的單字。 **start_offset**:`string` 單字相對於音訊起始的時間起始偏移。當 timestamp_granularities 包含 "word" 時出現。 **end_offset**:`string` 單字相對於音訊起始的時間結束偏移。當 timestamp_granularities 包含 "word" 時出現。 **speaker**:`string` 選用。此單字的講者標籤(例如 "spk_1"、"spk_2")。當 TranscriptionConfig 中設定了 diarization_mode 時出現。 **start_index**:`integer` 回應中歸屬此來源之區段的起始位置。索引表示區段的起始,以位元組為單位。 **end_index**:`integer` 歸屬區段的結束位置(不含該位置)。範例
文字
{
"type": "text",
"text": "Hello, how are you?"
}