L'appel de fonction vous permet de connecter des modèles à des outils et API externes. Au lieu de générer des réponses textuelles, le modèle détermine quand appeler des fonctions spécifiques et fournit les paramètres nécessaires pour exécuter des actions concrètes. Cela permet au modèle de servir de passerelle entre le langage naturel et les actions et données réelles. L'appel de fonctions présente trois principaux cas d'utilisation :
- Effectuer des actions : interagir avec des systèmes externes à l'aide d'API, par exemple pour planifier des rendez-vous, créer des factures, envoyer des e-mails ou contrôler des appareils domotiques.
- Augmenter les connaissances : accéder à des informations provenant de sources externes telles que des bases de données, des API et des bases de connaissances.
- Étendre les capacités : utilisez des outils externes pour effectuer des calculs et étendre les limites du modèle, par exemple en utilisant une calculatrice ou en créant des graphiques.
Vous trouverez ci-dessous des exemples de ces cas d'utilisation :
Planifier une réunion
Cet exemple montre comment définir une fonction qui planifie une réunion avec des participants à une heure spécifique, ce qui permet au modèle d'analyser les demandes des utilisateurs et de renvoyer des arguments structurés pour déclencher des actions dans des systèmes externes.
Python
from google import genai
schedule_meeting_function = {
"type": "function",
"name": "schedule_meeting",
"description": "Schedules a meeting with specified attendees at a given time and date.",
"parameters": {
"type": "object",
"properties": {
"attendees": {"type": "array", "items": {"type": "string"}},
"date": {"type": "string", "description": "Date (e.g., '2024-07-29')"},
"time": {"type": "string", "description": "Time (e.g., '15:00')"},
"topic": {"type": "string", "description": "The meeting topic."},
},
"required": ["attendees", "date", "time", "topic"],
},
}
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Schedule a meeting with Bob and Alice for 03/14/2025 at 10:00 AM about Q3 planning.",
tools=[{"type": "function", **schedule_meeting_function}],
)
for step in interaction.steps:
if step.type == "function_call":
print(f"Function to call: {step.name}")
print(f"Arguments: {step.arguments}")
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const scheduleMeetingFunction = {
type: 'function',
name: 'schedule_meeting',
description: 'Schedules a meeting with specified attendees at a given time and date.',
parameters: {
type: 'object',
properties: {
attendees: { type: 'array', items: { type: 'string' } },
date: { type: 'string', description: 'Date (e.g., "2024-07-29")' },
time: { type: 'string', description: 'Time (e.g., "15:00")' },
topic: { type: 'string', description: 'The meeting topic.' },
},
required: ['attendees', 'date', 'time', 'topic'],
},
};
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: 'Schedule a meeting with Bob and Alice for 03/27/2025 at 10:00 AM about Q3 planning.',
tools: [scheduleMeetingFunction],
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`Function to call: ${step.name}`);
console.log(`Arguments: ${JSON.stringify(step.arguments)}`);
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> attendeesProp = new HashMap<>();
attendeesProp.put("type", "array");
Map<String, Object> itemsMap = new HashMap<>(); itemsMap.put("type", "string"); attendeesProp.put("items", itemsMap);
Map<String, Object> dateProp = new HashMap<>();
dateProp.put("type", "string");
dateProp.put("description", "Date (e.g., \"2024-07-29\")");
Map<String, Object> timeProp = new HashMap<>();
timeProp.put("type", "string");
timeProp.put("description", "Time (e.g., \"15:00\")");
Map<String, Object> topicProp = new HashMap<>();
topicProp.put("type", "string");
topicProp.put("description", "The meeting topic.");
Map<String, Object> properties = new HashMap<>();
properties.put("attendees", attendeesProp);
properties.put("date", dateProp);
properties.put("time", timeProp);
properties.put("topic", topicProp);
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
parameters.put("properties", properties);
parameters.put("required", Arrays.asList("attendees", "date", "time", "topic"));
Function scheduleMeetingFunction =
Function.builder()
.name("schedule_meeting")
.description("Schedules a meeting with specified attendees at a given time and date.")
.parameters(parameters)
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Schedule a meeting with Bob and Alice for 03/27/2025 at 10:00 AM about Q3 planning."))
.tools(Arrays.asList(scheduleMeetingFunction))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep functionCall = (FunctionCallStep) step;
System.out.println("Function to call: " + functionCall.name().orElse(""));
System.out.println("Arguments: " + functionCall.arguments().orElse(null));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
// Define the function declaration for the model
scheduleMeetingFunc := &genai.FunctionDeclaration{
Name: "schedule_meeting",
Description: "Schedules a meeting with specified attendees at a given time and date.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"attendees": {
Type: genai.TypeArray,
Items: &genai.Schema{Type: genai.TypeString},
Description: "List of people attending the meeting.",
},
"date": {
Type: genai.TypeString,
Description: "Date (e.g., '2024-07-29')",
},
"time": {
Type: genai.TypeString,
Description: "Time (e.g., '15:00')",
},
"topic": {
Type: genai.TypeString,
Description: "The meeting topic.",
},
},
Required: []string{"attendees", "date", "time", "topic"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{scheduleMeetingFunc}},
},
}
// Send request with function declarations
response, err := client.Models.GenerateContent(
ctx,
"gemini-3.8-flash",
genai.Text("Schedule a meeting with Bob and Alice for 03/14/2025 at 10:00 AM about Q3 planning."),
config,
)
if err != nil {
log.Fatal(err)
}
// Check for a function call
if len(response.FunctionCalls()) > 0 {
functionCall := response.FunctionCalls()[0]
fmt.Printf("Function to call: %s\n", functionCall.Name)
fmt.Printf("Arguments: %v\n", functionCall.Args)
} else {
fmt.Println("No function call found in the response.")
fmt.Println(response.Text())
}
}
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",
"input": "Schedule a meeting with Bob and Alice for 03/27/2025 at 10:00 AM about Q3 planning.",
"tools": [{
"type": "function",
"name": "schedule_meeting",
"description": "Schedules a meeting with specified attendees at a given time and date.",
"parameters": {
"type": "object",
"properties": {
"attendees": {"type": "array", "items": {"type": "string"}},
"date": {"type": "string"},
"time": {"type": "string"},
"topic": {"type": "string"}
},
"required": ["attendees", "date", "time", "topic"]
}
}]
}'
Obtenir la météo
Cet exemple montre comment définir une fonction qui récupère les données de température pour un lieu, ce qui permet au modèle d'appeler des API externes pour répondre aux requêtes nécessitant des informations externes ou en temps réel.
Python
from google import genai
weather_function = {
"type": "function",
"name": "get_current_temperature",
"description": "Gets the current temperature for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city name, e.g. San Francisco",
},
},
"required": ["location"],
},
}
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="What's the temperature in London?",
tools=[weather_function],
)
for step in interaction.steps:
if step.type == "function_call":
print(f"Function to call: {step.name}")
print(f"Arguments: {step.arguments}")
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const weatherFunctionDeclaration = {
type: 'function',
name: 'get_current_temperature',
description: 'Gets the current temperature for a given location.',
parameters: {
type: 'object',
properties: {
location: {
type: 'string',
description: 'The city name, e.g. San Francisco',
},
},
required: ['location'],
},
};
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: "What's the temperature in London?",
tools: [weatherFunctionDeclaration],
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`Function to call: ${step.name}`);
console.log(`Arguments: ${JSON.stringify(step.arguments)}`);
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> locationProp = new HashMap<>();
locationProp.put("type", "string");
locationProp.put("description", "The city name, e.g. San Francisco");
Map<String, Object> properties = new HashMap<>();
properties.put("location", locationProp);
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
parameters.put("properties", properties);
parameters.put("required", Arrays.asList("location"));
Function weatherFunction =
Function.builder()
.name("get_current_temperature")
.description("Gets the current temperature for a given location.")
.parameters(parameters)
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("What's the temperature in London?"))
.tools(Arrays.asList(weatherFunction))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep functionCall = (FunctionCallStep) step;
System.out.println("Function to call: " + functionCall.name().orElse(""));
System.out.println("Arguments: " + functionCall.arguments().orElse(null));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
// Define the function declaration for the model
weatherFunc := &genai.FunctionDeclaration{
Name: "get_current_temperature",
Description: "Gets the current temperature for a given location.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"location": {
Type: genai.TypeString,
Description: "The city name, e.g. San Francisco",
},
},
Required: []string{"location"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{weatherFunc}},
},
}
// Send request with function declarations
response, err := client.Models.GenerateContent(
ctx,
"gemini-3.8-flash",
genai.Text("What's the temperature in London?"),
config,
)
if err != nil {
log.Fatal(err)
}
// Check for a function call
if len(response.FunctionCalls()) > 0 {
functionCall := response.FunctionCalls()[0]
fmt.Printf("Function to call: %s\n", functionCall.Name)
fmt.Printf("Arguments: %v\n", functionCall.Args)
} else {
fmt.Println("No function call found in the response.")
fmt.Println(response.Text())
}
}
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",
"input": "What'\''s the temperature in London?",
"tools": [{
"type": "function",
"name": "get_current_temperature",
"description": "Gets the current temperature for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city name"}
},
"required": ["location"]
}
}]
}'
Créer un graphique
Cet exemple montre comment définir une fonction qui génère un graphique à barres à partir de données structurées. Il illustre la façon dont le modèle peut utiliser des outils externes pour effectuer des calculs ou créer des éléments visuels :
Python
from google import genai
create_chart_function = {
"type": "function",
"name": "create_bar_chart",
"description": "Creates a bar chart given a title, labels, and values.",
"parameters": {
"type": "object",
"properties": {
"title": {"type": "string", "description": "The title for the chart."},
"labels": {"type": "array", "items": {"type": "string"}},
"values": {"type": "array", "items": {"type": "number"}},
},
"required": ["title", "labels", "values"],
},
}
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Create a bar chart titled 'Quarterly Sales' with Q1: 50000, Q2: 75000, Q3: 60000.",
tools=[create_chart_function],
)
for step in interaction.steps:
if step.type == "function_call":
print(f"Function to call: {step.name}")
print(f"Arguments: {step.arguments}")
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const createChartFunctionDeclaration = {
type: 'function',
name: 'create_bar_chart',
description: 'Creates a bar chart given a title, labels, and values.',
parameters: {
type: 'object',
properties: {
title: { type: 'string', description: 'The title for the chart.' },
labels: { type: 'array', items: { type: 'string' } },
values: { type: 'array', items: { type: 'number' } },
},
required: ['title', 'labels', 'values'],
},
};
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: "Create a bar chart titled 'Quarterly Sales' with Q1: 50000, Q2: 75000, Q3: 60000.",
tools: [createChartFunctionDeclaration],
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`${step.name}(${JSON.stringify(step.arguments)})`);
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> properties = new HashMap<>();
Map<String, Object> titleMap = new HashMap<>(); titleMap.put("type", "string"); titleMap.put("description", "The title for the chart."); properties.put("title", titleMap);
Map<String, Object> labelsMap = new HashMap<>(); labelsMap.put("type", "array"); labelsMap.put("items", Collections.singletonMap("type", "string")); properties.put("labels", labelsMap);
Map<String, Object> valuesMap = new HashMap<>(); valuesMap.put("type", "array"); valuesMap.put("items", Collections.singletonMap("type", "number")); properties.put("values", valuesMap);
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
parameters.put("properties", properties);
parameters.put("required", Arrays.asList("title", "labels", "values"));
Function createChartFunction =
Function.builder()
.name("create_bar_chart")
.description("Creates a bar chart given a title, labels, and values.")
.parameters(parameters)
.build();
CreateModelInteraction params =
CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Create a bar chart titled 'Quarterly Sales' with Q1: 50000, Q2: 75000, Q3: 60000."))
.tools(Arrays.asList(createChartFunction))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep functionCall = (FunctionCallStep) step;
System.out.println(functionCall.name().orElse("") + "(" + functionCall.arguments().orElse(null) + ")");
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
// Define the function declaration for the model
createChartFunc := &genai.FunctionDeclaration{
Name: "create_bar_chart",
Description: "Creates a bar chart given a title, labels, and values.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"title": {
Type: genai.TypeString,
Description: "The title for the chart.",
},
"labels": {
Type: genai.TypeArray,
Items: &genai.Schema{Type: genai.TypeString},
},
"values": {
Type: genai.TypeArray,
Items: &genai.Schema{Type: genai.TypeNumber},
},
},
Required: []string{"title", "labels", "values"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{createChartFunc}},
},
}
// Send request with function declarations
response, err := client.Models.GenerateContent(
ctx,
"gemini-3.8-flash",
genai.Text("Create a bar chart titled 'Quarterly Sales' with Q1: 50000, Q2: 75000, Q3: 60000."),
config,
)
if err != nil {
log.Fatal(err)
}
// Check for a function call
if len(response.FunctionCalls()) > 0 {
functionCall := response.FunctionCalls()[0]
fmt.Printf("Function to call: %s\n", functionCall.Name)
fmt.Printf("Arguments: %v\n", functionCall.Args)
} else {
fmt.Println("No function call found in the response.")
fmt.Println(response.Text())
}
}
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",
"input": "Create a bar chart titled '\''Quarterly Sales'\'' with Q1: 50000, Q2: 75000, Q3: 60000.",
"tools": [{
"type": "function",
"name": "create_bar_chart",
"description": "Creates a bar chart given a title, labels, and values.",
"parameters": {
"type": "object",
"properties": {
"title": {"type": "string"},
"labels": {"type": "array", "items": {"type": "string"}},
"values": {"type": "array", "items": {"type": "number"}}
},
"required": ["title", "labels", "values"]
}
}]
}'
Fonctionnement des appels de fonction
L'appel de fonction implique une interaction structurée entre votre application, le modèle et les fonctions externes :
- Définir la déclaration de fonction : définissez le nom, les paramètres et l'objectif de la fonction pour le modèle.
- Appeler le LLM avec des déclarations de fonctions : envoyez la requête de l'utilisateur au modèle, ainsi que la ou les déclarations de fonctions.
- Exécution du code de la fonction (votre responsabilité) : le modèle n'exécute pas la fonction lui-même. Extrayez le nom et les arguments, puis exécutez-les dans votre application.
- Créer une réponse conviviale : renvoyez le résultat au modèle pour obtenir une réponse finale et conviviale.
Ce processus peut être répété sur plusieurs tours. Le modèle permet d'appeler plusieurs fonctions en un seul tour (appels de fonction en parallèle) et de manière séquentielle (appels de fonction compositionnels).
Étape 1 : Définir une déclaration de fonction
Python
set_light_values_declaration = {
"type": "function",
"name": "set_light_values",
"description": "Sets the brightness and color temperature of a light.",
"parameters": {
"type": "object",
"properties": {
"brightness": {
"type": "integer",
"description": "Light level from 0 to 100",
},
"color_temp": {
"type": "string",
"enum": ["daylight", "cool", "warm"],
"description": "Color temperature",
},
},
"required": ["brightness", "color_temp"],
},
}
def set_light_values(brightness: int, color_temp: str) -> dict:
"""Set the brightness and color temperature of a room light."""
return {"brightness": brightness, "colorTemperature": color_temp}
JavaScript
const setLightValuesTool = {
type: 'function',
name: 'set_light_values',
description: 'Sets the brightness and color temperature of a light.',
parameters: {
type: 'object',
properties: {
brightness: { type: 'number', description: 'Light level from 0 to 100' },
color_temp: { type: 'string', enum: ['daylight', 'cool', 'warm'] },
},
required: ['brightness', 'color_temp'],
},
};
function setLightValues(brightness, color_temp) {
return { brightness: brightness, colorTemperature: color_temp };
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import "google.golang.org/genai"
var setLightValuesDeclaration = &genai.FunctionDeclaration{
Name: "set_light_values",
Description: "Sets the brightness and color temperature of a light.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"brightness": {
Type: genai.TypeInteger,
Description: "Light level from 0 to 100",
},
"color_temp": {
Type: genai.TypeString,
Enum: []string{"daylight", "cool", "warm"},
Description: "Color temperature",
},
},
Required: []string{"brightness", "color_temp"},
},
}
func setLightValues(brightness int, colorTemp string) map[string]any {
return map[string]any{"brightness": brightness, "colorTemperature": colorTemp}
}
Étape 2 : Appeler le modèle avec les déclarations de fonction
Python
from google import genai
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Turn the lights down to a romantic level",
tools=[set_light_values_declaration],
)
fc_step = next(s for s in interaction.steps if s.type == "function_call")
print(fc_step)
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: 'Turn the lights down to a romantic level',
tools: [setLightValuesTool],
});
const fcStep = interaction.steps.find(s => s.type === 'function_call');
console.log(fcStep);
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{setLightValuesDeclaration}},
},
}
contents := []*genai.Content{
genai.NewContentFromText("Turn the lights down to a romantic level", genai.RoleUser),
}
response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, config)
if err != nil {
log.Fatal(err)
}
fmt.Println(response.FunctionCalls()[0])
Le modèle renvoie une étape function_call avec type, name et arguments :
type='function_call'
name='set_light_values'
arguments={'color_temp': 'warm', 'brightness': 25}
Étape 3 : Exécutez la fonction
Python
fc_step = next(s for s in interaction.steps if s.type == "function_call")
if fc_step.name == "set_light_values":
result = set_light_values(**fc_step.arguments)
print(f"Function execution result: {result}")
JavaScript
const fcStep = interaction.steps.find(s => s.type === 'function_call');
let result;
if (fcStep.name === 'set_light_values') {
result = setLightValues(fcStep.arguments.brightness, fcStep.arguments.color_temp);
console.log(`Function execution result: ${JSON.stringify(result)}`);
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
toolCall := response.FunctionCalls()[0]
var result map[string]any
if toolCall.Name == "set_light_values" {
brightness := int(toolCall.Args["brightness"].(float64))
colorTemp := toolCall.Args["color_temp"].(string)
result = setLightValues(brightness, colorTemp)
fmt.Printf("Function execution result: %v\n", result)
}
Étape 4 : Renvoyer le résultat au modèle
Python
final_interaction = client.interactions.create(
model="gemini-3.8-flash",
input=[
{
"type": "function_result",
"name": fc_step.name,
"call_id": fc_step.id,
"result": [{"type": "text", "text": json.dumps(result)}],
}
],
tools=[set_light_values_declaration],
previous_interaction_id=interaction.id,
)
print(final_interaction.output_text)
JavaScript
const finalInteraction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: [{
type: 'function_result',
name: fcStep.name,
call_id: fcStep.id,
result: [{ type: 'text', text: JSON.stringify(result) }]
}],
tools: [setLightValuesTool],
previous_interaction_id: interaction.id,
});
console.log(finalInteraction.output_text);
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
functionResponsePart := &genai.Part{
FunctionResponse: &genai.FunctionResponse{
ID: toolCall.ID,
Name: toolCall.Name,
Response: result,
},
}
contents = append(contents, response.Candidates[0].Content)
contents = append(contents, &genai.Content{
Role: genai.RoleUser,
Parts: []*genai.Part{functionResponsePart},
})
finalResponse, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, config)
if err != nil {
log.Fatal(err)
}
fmt.Println(finalResponse.Text())
Appel de fonction sans état
Vous pouvez également utiliser les appels de fonction en mode sans état en gérant l'historique des conversations côté client et en définissant store=false.
En mode sans état, vous devez transmettre l'intégralité de l'historique de la conversation dans le champ input de chaque requête suivante. Cet historique doit inclure :
1. Étape user_input initiale.
2. Toutes les étapes générées par le modèle renvoyées au tour 1 (y compris les étapes thought et function_call) exactement telles qu'elles ont été reçues.
3. Étape function_result contenant le résultat de votre fonction exécutée.
Python
from google import genai
import json
client = genai.Client()
history = [
{
"type": "user_input",
"content": [{"type": "text", "text": "Turn the lights down to a romantic level"}]
}
]
interaction = client.interactions.create(
model="gemini-3.8-flash",
store=False,
input=history,
tools=[set_light_values_declaration],
)
for step in interaction.steps:
history.append(step.model_dump())
fc_step = next(s for s in interaction.steps if s.type == "function_call")
if fc_step.name == "set_light_values":
result = set_light_values(**fc_step.arguments)
history.append({
"type": "function_result",
"name": fc_step.name,
"call_id": fc_step.id,
"result": [{"type": "text", "text": json.dumps(result)}],
})
final_interaction = client.interactions.create(
model="gemini-3.8-flash",
store=False,
input=history,
tools=[set_light_values_declaration],
)
print(final_interaction.output_text)
JavaScript
import { GoogleGenAI } from "@google/genai";
const client = new GoogleGenAI({});
async function main() {
const history = [
{
type: "user_input",
content: [{ type: "text", text: "Turn the lights down to a romantic level" }]
}
];
const interaction = await client.interactions.create({
model: "gemini-3.8-flash",
store: false,
input: history,
tools: [setLightValuesTool],
});
history.push(...interaction.steps);
const fcStep = interaction.steps.find(s => s.type === 'function_call');
let result;
if (fcStep.name === 'set_light_values') {
result = setLightValues(fcStep.arguments.brightness, fcStep.arguments.color_temp);
}
history.push({
type: 'function_result',
name: fcStep.name,
call_id: fcStep.id,
result: [{ type: 'text', text: JSON.stringify(result) }]
});
const finalInteraction = await client.interactions.create({
model: 'gemini-3.8-flash',
store: false,
input: history,
tools: [setLightValuesTool],
});
console.log(finalInteraction.output_text);
}
await main();
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{setLightValuesDeclaration}},
},
}
history := []*genai.Content{
genai.NewContentFromText("Turn the lights down to a romantic level", genai.RoleUser),
}
response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", history, config)
if err != nil {
log.Fatal(err)
}
toolCall := response.FunctionCalls()[0]
brightness := int(toolCall.Args["brightness"].(float64))
colorTemp := toolCall.Args["color_temp"].(string)
result := setLightValues(brightness, colorTemp)
history = append(history, response.Candidates[0].Content)
history = append(history, &genai.Content{
Role: genai.RoleUser,
Parts: []*genai.Part{
{
FunctionResponse: &genai.FunctionResponse{
ID: toolCall.ID,
Name: toolCall.Name,
Response: result,
},
},
},
})
finalResponse, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", history, config)
if err != nil {
log.Fatal(err)
}
fmt.Println(finalResponse.Text())
REST
# Turn 1: Send request with tools and store: false
RESPONSE1=$(curl -s -X POST "https://generativelanguage.googleapis.com/v1beta/interactions" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "gemini-3.8-flash",
"store": false,
"input": [
{
"type": "user_input",
"content": "Turn the lights down to a romantic level"
}
],
"tools": [{
"type": "function",
"name": "set_light_values",
"description": "Sets the brightness and color temperature of a light.",
"parameters": {
"type": "object",
"properties": {
"brightness": {"type": "integer", "description": "Light level from 0 to 100"},
"color_temp": {"type": "string", "enum": ["daylight", "cool", "warm"]}
},
"required": ["brightness", "color_temp"]
}
}]
}')
# Extract model steps (thought, function_call)
MODEL_STEPS=$(echo "$RESPONSE1" | jq '.steps')
# Extract function call details to execute
FC_NAME=$(echo "$RESPONSE1" | jq -r '.steps[] | select(.type=="function_call") | .name')
FC_ID=$(echo "$RESPONSE1" | jq -r '.steps[] | select(.type=="function_call") | .id')
# Assume local execution returns: {"brightness": 25, "colorTemperature": "warm"}
RESULT="{\"brightness\": 25, \"colorTemperature\": \"warm\"}"
# Reconstruct history for Turn 2
HISTORY=$(jq -n \
--argjson first_input '[{"type": "user_input", "content": "Turn the lights down to a romantic level"}]' \
--argjson model_steps "$MODEL_STEPS" \
--arg fc_name "$FC_NAME" \
--arg fc_id "$FC_ID" \
--arg result "$RESULT" \
'$first_input + $model_steps + [{"type": "function_result", "name": $fc_name, "call_id": $fc_id, "result": [{"type": "text", "text": $result}]}]')
# Turn 2: Send the full history
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\",
\"store\": false,
\"input\": $HISTORY,
\"tools\": [{
\"type\": \"function\",
\"name\": \"set_light_values\",
\"description\": \"Sets the brightness and color temperature of a light.\",
\"parameters\": {
\"type\": \"object\",
\"properties\": {
\"brightness\": {\"type\": \"integer\"},
\"color_temp\": {\"type\": \"string\"}
},
\"required\": [\"brightness\", \"color_temp\"]
}
}]
}"
Déclarations de fonctions
Une déclaration de fonction est transmise en tant qu'outil et inclut les éléments suivants :
type(chaîne) : doit être"function"pour les fonctions personnalisées.name(chaîne) : nom de fonction unique (utilisez des traits de soulignement ou la casse mixte).description(chaîne) : explication claire de l'objectif de la fonction.parameters(objet) : paramètres d'entrée attendus par la fonction.type(chaîne) : type de données global, tel queobject.properties(objet) : paramètres individuels avec type et description.required(tableau) : noms des paramètres obligatoires.
Appel de fonction avec des modèles à raisonnement
Les modèles de la série Gemini 3 utilisent un processus de réflexion interne qui améliore l'appel de fonction. Les SDK gèrent automatiquement les signatures de pensée pour vous.
Appel de fonction en parallèle
Appelez plusieurs fonctions à la fois lorsqu'elles sont indépendantes :
Python
power_disco_ball = {"type": "function", "name": "power_disco_ball", "description": "Powers the disco ball.",
"parameters": {"type": "object", "properties": {"power": {"type": "boolean"}}, "required": ["power"]}}
start_music = {"type": "function", "name": "start_music", "description": "Play music.",
"parameters": {"type": "object", "properties": {"energetic": {"type": "boolean"}, "loud": {"type": "boolean"}}, "required": ["energetic", "loud"]}}
dim_lights = {"type": "function", "name": "dim_lights", "description": "Dim the lights.",
"parameters": {"type": "object", "properties": {"brightness": {"type": "number"}}, "required": ["brightness"]}}
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Turn this place into a party!",
tools=[power_disco_ball, start_music, dim_lights],
generation_config={"tool_choice": "any"},
)
for step in interaction.steps:
if step.type == "function_call":
args = ", ".join(f"{key}={val}" for key, val in step.arguments.items())
print(f"{step.name}({args})")
JavaScript
const powerDiscoBall = { type: 'function', name: 'power_disco_ball', description: 'Powers the disco ball.',
parameters: { type: 'object', properties: { power: { type: 'boolean' } }, required: ['power'] } };
const startMusic = { type: 'function', name: 'start_music', description: 'Play music.',
parameters: { type: 'object', properties: { energetic: { type: 'boolean' }, loud: { type: 'boolean' } }, required: ['energetic', 'loud'] } };
const dimLights = { type: 'function', name: 'dim_lights', description: 'Dim the lights.',
parameters: { type: 'object', properties: { brightness: { type: 'number' } }, required: ['brightness'] } };
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: 'Turn this place into a party!',
tools: [powerDiscoBall, startMusic, dimLights],
generation_config: { tool_choice: 'any' },
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`${step.name}(${JSON.stringify(step.arguments)})`);
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
powerDiscoBall := &genai.FunctionDeclaration{
Name: "power_disco_ball",
Description: "Powers the disco ball.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"power": {Type: genai.TypeBoolean},
},
Required: []string{"power"},
},
}
startMusic := &genai.FunctionDeclaration{
Name: "start_music",
Description: "Play music.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"energetic": {Type: genai.TypeBoolean},
"loud": {Type: genai.TypeBoolean},
},
Required: []string{"energetic", "loud"},
},
}
dimLights := &genai.FunctionDeclaration{
Name: "dim_lights",
Description: "Dim the lights.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"brightness": {Type: genai.TypeNumber},
},
Required: []string{"brightness"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{powerDiscoBall, startMusic, dimLights}},
},
ToolConfig: &genai.ToolConfig{
FunctionCallingConfig: &genai.FunctionCallingConfig{
Mode: genai.FunctionCallingConfigModeAny,
},
},
}
response, err := client.Models.GenerateContent(
ctx,
"gemini-3.8-flash",
genai.Text("Turn this place into a party!"),
config,
)
if err != nil {
log.Fatal(err)
}
for _, fn := range response.FunctionCalls() {
fmt.Printf("%s(%v)\n", fn.Name, fn.Args)
}
}
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",
"input": "Turn this place into a party!",
"tools": [
{
"type": "function",
"name": "power_disco_ball",
"description": "Powers the disco ball.",
"parameters": {
"type": "object",
"properties": {
"power": {"type": "boolean"}
},
"required": ["power"]
}
},
{
"type": "function",
"name": "start_music",
"description": "Play music.",
"parameters": {
"type": "object",
"properties": {
"energetic": {"type": "boolean"},
"loud": {"type": "boolean"}
},
"required": ["energetic", "loud"]
}
},
{
"type": "function",
"name": "dim_lights",
"description": "Dim the lights.",
"parameters": {
"type": "object",
"properties": {
"brightness": {"type": "number"}
},
"required": ["brightness"]
}
}
]
}'
Appel de fonction compositionnel
Associez plusieurs appels de fonction pour les requêtes complexes (par exemple, obtenez d'abord la position, puis la météo pour cette position).
Python
get_weather_forecast_declaration = {
"type": "function",
"name": "get_weather_forecast",
"description": "Gets the current weather temperature for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The location"},
},
"required": ["location"],
},
}
set_thermostat_temperature_declaration = {
"type": "function",
"name": "set_thermostat_temperature",
"description": "Sets the thermostat to a desired temperature.",
"parameters": {
"type": "object",
"properties": {
"temperature": {
"type": "integer",
"description": "The temperature in Celsius",
},
},
"required": ["temperature"],
},
}
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="If it's warmer than 20°C in London, set the thermostat to 20°C, otherwise 18°C.",
tools=[
get_weather_forecast_declaration,
set_thermostat_temperature_declaration,
],
)
for step in interaction.steps:
if step.type == "function_call":
print(f"Function to call: {step.name}")
print(f"Arguments: {step.arguments}")
elif hasattr(step, "content") and step.content:
for part in step.content:
if hasattr(part, "text"):
print(part.text)
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const getWeatherForecastTool = {
type: 'function',
name: 'get_weather_forecast',
description: 'Gets the current weather temperature for a given location.',
parameters: {
type: 'object',
properties: {
location: { type: 'string', description: 'The location' },
},
required: ['location'],
},
};
const setThermostatTemperatureTool = {
type: 'function',
name: 'set_thermostat_temperature',
description: 'Sets the thermostat to a desired temperature.',
parameters: {
type: 'object',
properties: {
temperature: {
type: 'integer',
description: 'The temperature in Celsius',
},
},
required: ['temperature'],
},
};
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: "If it's warmer than 20°C in London, set the thermostat to 20°C, otherwise 18°C.",
tools: [
getWeatherForecastTool,
setThermostatTemperatureTool,
],
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`Function to call: ${step.name}`);
console.log(`Arguments: ${JSON.stringify(step.arguments)}`);
} else if (step.content) {
for (const part of step.content) {
if (part.text) {
console.log(part.text);
}
}
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
getWeatherForecastDecl := &genai.FunctionDeclaration{
Name: "get_weather_forecast",
Description: "Gets the current weather temperature for a given location.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"location": {Type: genai.TypeString, Description: "The location"},
},
Required: []string{"location"},
},
}
setThermostatTemperatureDecl := &genai.FunctionDeclaration{
Name: "set_thermostat_temperature",
Description: "Sets the thermostat to a desired temperature.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"temperature": {Type: genai.TypeInteger, Description: "The temperature in Celsius"},
},
Required: []string{"temperature"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{getWeatherForecastDecl, setThermostatTemperatureDecl}},
},
}
response, err := client.Models.GenerateContent(
ctx,
"gemini-3.8-flash",
genai.Text("If it's warmer than 20°C in London, set the thermostat to 20°C, otherwise 18°C."),
config,
)
if err != nil {
log.Fatal(err)
}
for _, fn := range response.FunctionCalls() {
fmt.Printf("Function to call: %s\n", fn.Name)
fmt.Printf("Arguments: %v\n", fn.Args)
}
}
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",
"input": "If it'\''s warmer than 20°C in London, set the thermostat to 20°C, otherwise 18°C.",
"tools": [
{
"type": "function",
"name": "get_weather_forecast",
"description": "Gets the current weather temperature for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string"}
},
"required": ["location"]
}
},
{
"type": "function",
"name": "set_thermostat_temperature",
"description": "Sets the thermostat to a desired temperature.",
"parameters": {
"type": "object",
"properties": {
"temperature": {"type": "integer"}
},
"required": ["temperature"]
}
}
]
}'
Modes d'appel de fonction
Contrôlez la façon dont le modèle utilise les outils à l'aide de tool_choice dans generation_config :
auto(par défaut) : le modèle décide s'il faut appeler une fonction ou répondre directement.any: le modèle est contraint de toujours prédire un appel de fonction.none: le modèle n'est pas autorisé à effectuer des appels de fonction.validated: le modèle garantit le respect du schéma de la fonction.
Python
generation_config = {
"tool_choice": {
"allowed_tools": {
"mode": "any",
"tools": ["get_current_temperature"]
}
}
}
JavaScript
const generation_config = {
tool_choice: {
allowed_tools: {
mode: 'any',
tools: ['get_current_temperature']
}
}
};
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
// Configure function calling mode
toolConfig := &genai.ToolConfig{
FunctionCallingConfig: &genai.FunctionCallingConfig{
Mode: genai.FunctionCallingConfigModeAny,
AllowedFunctionNames: []string{"get_current_temperature"},
},
}
// Create the generation config
config := &genai.GenerateContentConfig{
Tools: tools, // not defined here.
ToolConfig: toolConfig,
}
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",
"input": "What is the temperature in Boston?",
"tools": [{
"type": "function",
"name": "get_current_temperature",
"description": "Gets the current temperature for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string"}
},
"required": ["location"]
}
}],
"generation_config": {
"tool_choice": {
"allowed_tools": {
"mode": "any",
"tools": ["get_current_temperature"]
}
}
}
}'
Utilisation de plusieurs outils
Vous pouvez activer plusieurs outils, en combinant des outils intégrés avec l'appel de fonction dans la même requête. Les modèles Gemini 3 peuvent combiner des outils intégrés avec l'appel de fonction prêt à l'emploi dans les interactions. La transmission de previous_interaction_id fait automatiquement circuler le contexte de l'outil intégré.
Python
from google import genai
import json
client = genai.Client()
get_weather = {
"type": "function",
"name": "get_weather",
"description": "Gets the weather for a requested city.",
"parameters": {
"type": "object",
"properties": {
"city": {
"type": "string",
"description": "The city and state, e.g. Utqiaġvik, Alaska",
},
},
"required": ["city"],
},
}
tools = [
{"type": "google_search"},
get_weather
]
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="What is the northernmost city in the United States? What's the weather like there today?",
tools=tools
)
for step in interaction.steps:
if step.type == "function_call":
print(f"Function call: {step.name} (ID: {step.id})")
result = {"response": "Very cold. 22 degrees Fahrenheit."}
interaction_2 = client.interactions.create(
model="gemini-3.8-flash",
previous_interaction_id=interaction.id,
tools=tools,
input=[{
"type": "function_result",
"name": step.name,
"call_id": step.id,
"result": [{"type": "text", "text": json.dumps(result)}]
}]
)
print(interaction_2.output_text)
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const weatherTool = {
type: 'function',
name: 'get_weather',
description: 'Gets the weather for a given location.',
parameters: {
type: 'object',
properties: {
location: {
type: 'string',
description: 'The city and state, e.g. San Francisco, CA',
},
},
required: ['location'],
},
};
const tools = [
{ type: 'google_search' }, // Built-in tool
weatherTool,
];
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: "What is the northernmost city in the United States? What's the weather like there today?",
tools: tools,
});
for (const step of interaction.steps) {
if (step.type === 'function_call') {
console.log(`Function call: ${step.name} (ID: ${step.id})`);
const result = { response: 'Very cold. 22 degrees Fahrenheit.' };
const interaction_2 = await client.interactions.create({
model: 'gemini-3.8-flash',
previous_interaction_id: interaction.id,
tools: tools,
input: [
{
type: 'function_result',
name: step.name,
call_id: step.id,
result: [{ type: 'text', text: JSON.stringify(result) }],
},
],
});
console.log(interaction_2.output_text);
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
getWeather := &genai.FunctionDeclaration{
Name: "get_weather",
Description: "Gets the weather for a given location.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"location": {
Type: genai.TypeString,
Description: "The city and state, e.g. San Francisco, CA",
},
},
Required: []string{"location"},
},
}
tools := []*genai.Tool{
{GoogleSearch: &genai.GoogleSearch{}},
{FunctionDeclarations: []*genai.FunctionDeclaration{getWeather}},
}
config := &genai.GenerateContentConfig{
Tools: tools,
}
prompt := "What is the northernmost city in the United States? What's the weather like there today?"
response1, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", genai.Text(prompt), config)
if err != nil {
log.Fatal(err)
}
toolCall := response1.FunctionCalls()[0]
fmt.Printf("Function call: %s (ID: %s)\n", toolCall.Name, toolCall.ID)
history := []*genai.Content{
genai.NewContentFromText(prompt, genai.RoleUser),
response1.Candidates[0].Content,
{
Role: genai.RoleUser,
Parts: []*genai.Part{
{
FunctionResponse: &genai.FunctionResponse{
ID: toolCall.ID,
Name: toolCall.Name,
Response: map[string]any{"response": "Very cold. 22 degrees Fahrenheit."},
},
},
},
},
}
response2, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", history, config)
if err != nil {
log.Fatal(err)
}
fmt.Println(response2.Text())
}
REST
# Turn 1: Send request with built-in google_search tool and custom weather tool
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": "What is the northernmost city in the United States? What'\''s the weather like there today?",
"tools": [
{"type": "google_search"},
{
"type": "function",
"name": "get_weather",
"description": "Gets the weather for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"}
},
"required": ["location"]
}
}
]
}'
# Turn 2: Provide function result and pass previous_interaction_id
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",
"previous_interaction_id": "INTERACTION_ID",
"tools": [
{"type": "google_search"},
{
"type": "function",
"name": "get_weather",
"description": "Gets the weather for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"}
},
"required": ["location"]
}
}
],
"input": [
{
"type": "function_result",
"name": "get_weather",
"call_id": "call_123",
"result": [{"type": "text", "text": "{\"response\": \"Very cold. 22 degrees Fahrenheit.\"}"}]
}
]
}'
Réponses de fonction multimodales
Pour les modèles de la série Gemini 3, vous pouvez inclure du contenu multimodal dans les parties de réponse de fonction que vous envoyez au modèle. Le modèle peut traiter ce contenu multimodal lors de son prochain tour pour produire une réponse plus pertinente.
Pour inclure des données multimodales dans une réponse de fonction, ajoutez-les sous forme d'un ou plusieurs blocs de contenu dans le champ result de l'étape function_result. Chaque bloc de contenu doit spécifier son type (par exemple, "text", "image").
L'exemple suivant montre comment renvoyer une réponse de fonction contenant des données d'image au modèle lors d'une interaction :
Python
import base64
from google import genai
import requests
client = genai.Client()
tool_call = next(s for s in interaction.steps if s.type == "function_call")
image_path = "https://goo.gle/instrument-img"
image_bytes = requests.get(image_path).content
base64_image_data = base64.b64encode(image_bytes).decode("utf-8")
final_interaction = client.interactions.create(
model="gemini-3.8-flash",
previous_interaction_id=interaction.id,
input=[
{
"type": "function_result",
"name": tool_call.name,
"call_id": tool_call.id,
"result": [
{"type": "text", "text": "instrument.jpg"},
{
"type": "image",
"mime_type": "image/jpeg",
"data": base64_image_data,
},
],
}
],
)
print(final_interaction.output_text)
JavaScript
import { GoogleGenAI } from "@google/genai";
const client = new GoogleGenAI({});
const toolCall = interaction.steps.find(s => s.type === 'function_call');
const base64ImageData = "BASE64_IMAGE_DATA";
const finalInteraction = await client.interactions.create({
model: 'gemini-3.8-flash',
previous_interaction_id: interaction.id,
input: [{
type: 'function_result',
name: toolCall.name,
call_id: toolCall.id,
result: [
{ type: 'text', text: 'instrument.jpg' },
{
type: 'image',
mime_type: 'image/jpeg',
data: base64ImageData,
}
]
}]
});
console.log(finalInteraction.output_text);
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import (
"context"
"fmt"
"io"
"log"
"net/http"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
// 1. Define the function tool
getImageDeclaration := &genai.FunctionDeclaration{
Name: "get_image",
Description: "Retrieves the image file reference for a specific order item.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"item_name": {
Type: genai.TypeString,
Description: "The name or description of the item ordered (e.g., 'instrument').",
},
},
Required: []string{"item_name"},
},
}
tools := []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{getImageDeclaration}},
}
// 2. Send a message that triggers the tool
prompt := "Show me the instrument I ordered last month."
response1, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", genai.Text(prompt), &genai.GenerateContentConfig{
Tools: tools,
})
if err != nil {
log.Fatal(err)
}
// 3. Handle the function call
functionCall := response1.FunctionCalls()[0]
requestedItem := functionCall.Args["item_name"]
fmt.Printf("Model wants to call: %s\n", functionCall.Name)
fmt.Printf("Calling external tool for: %v\n", requestedItem)
resp, err := http.Get("https://goo.gle/instrument-img")
if err != nil {
log.Fatal(err)
}
defer resp.Body.Close()
imageBytes, err := io.ReadAll(resp.Body)
if err != nil {
log.Fatal(err)
}
functionResponseData := map[string]any{
"image_ref": map[string]any{"$ref": "instrument.jpg"},
}
functionResponseMultimodalData := &genai.FunctionResponsePart{
InlineData: &genai.FunctionResponseBlob{
MIMEType: "image/jpeg",
DisplayName: "instrument.jpg",
Data: imageBytes,
},
}
// 4. Send the tool's result back
history := []*genai.Content{
genai.NewContentFromText(prompt, genai.RoleUser),
response1.Candidates[0].Content,
{
Role: genai.RoleUser,
Parts: []*genai.Part{
{
FunctionResponse: &genai.FunctionResponse{
ID: functionCall.ID,
Name: functionCall.Name,
Response: functionResponseData,
Parts: []*genai.FunctionResponsePart{functionResponseMultimodalData},
},
},
},
},
}
response2, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", history, &genai.GenerateContentConfig{
Tools: tools,
ThinkingConfig: &genai.ThinkingConfig{
IncludeThoughts: true,
},
})
if err != nil {
log.Fatal(err)
}
fmt.Printf("\nFinal model response: %s\n", response2.Text())
}
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",
"previous_interaction_id": "INTERACTION_ID",
"input": [
{
"type": "function_result",
"name": "get_image",
"call_id": "call_123",
"result": [
{"type": "text", "text": "instrument.jpg"},
{
"type": "image",
"mime_type": "image/jpeg",
"data": "BASE64_IMAGE_DATA"
}
]
}
]
}'
Appel de fonction avec sortie structurée
Pour les modèles de la série Gemini 3, combinez l'appel de fonction avec la sortie structurée pour obtenir des réponses dont le format est cohérent.
MCP (Model Context Protocol) à distance
L'API Interactions permet de se connecter à des serveurs MCP distants pour donner au modèle l'accès à des outils et services externes. Vous fournissez les name et url du serveur dans la configuration des outils.
Lorsque vous utilisez Remote MCP, tenez compte des contraintes suivantes :
- Types de serveur : le serveur MCP à distance ne fonctionne qu'avec les serveurs HTTP diffusable. Les serveurs SSE (Server-Sent Events) ne sont pas compatibles.
- Nommage : les noms de serveur MCP ne doivent pas inclure le caractère
-. Utilisez plutôt les noms de serveursnake_case.
| Champ | Type | Obligatoire | Description |
|---|---|---|---|
type |
string |
Oui | doit être "mcp_server" |
name |
string |
Non | Nom à afficher du serveur MCP. |
url |
string |
Non | URL complète du point de terminaison du serveur MCP. |
headers |
object |
Non | Paires clé-valeur envoyées en tant qu'en-têtes HTTP avec chaque requête au serveur (par exemple, jetons d'authentification). |
allowed_tools |
array |
Non | Limitez les outils du serveur que l'agent peut appeler. |
Exemple
Python
from google import genai
client = genai.Client()
interaction = client.interactions.create(
model="gemini-3.8-flash",
input="Check the weather in San Francisco.",
tools=[
{
"type": "mcp_server",
"name": "weather",
"url": "https://gemini-api-demos.uc.r.appspot.com/mcp",
}
]
)
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const interaction = await client.interactions.create({
model: 'gemini-3.8-flash',
input: 'Check the weather in San Francisco.',
tools: [
{
type: 'mcp_server',
name: 'weather',
url: 'https://gemini-api-demos.uc.r.appspot.com/mcp'
}
]
});
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
REST
curl -X POST "https://generativelanguage.googleapis.com/v1beta/interactions" \
-H "Content-Type: application/json" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-d '{
"model": "gemini-3.8-flash",
"input": "Check the weather in San Francisco.",
"tools": [
{
"type": "mcp_server",
"name": "weather",
"url": "https://gemini-api-demos.uc.r.appspot.com/mcp"
}
]
}'
Diffuser des appels d'outils
Lorsque vous utilisez des outils avec le streaming, le modèle génère des appels de fonction sous forme de séquence d'événements step.delta dans le flux. Les arguments d'outil peuvent être diffusés en tant qu'arguments partiels à l'aide de arguments. Vous devez agréger ces deltas pour reconstruire les appels d'outils complets avant de les exécuter.
Python
import json
from google import genai
client = genai.Client()
weather_tool = {
"type": "function",
"name": "get_weather",
"description": "Gets the weather for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state"}
},
"required": ["location"]
}
}
stream = client.interactions.create(
model="gemini-3.8-flash",
input="What is the weather in Paris?",
tools=[weather_tool],
stream=True
)
current_calls = {}
tool_calls = []
for event in stream:
if event.event_type == "step.start":
if event.step.type == "function_call":
current_calls[event.index] = {
"id": event.step.id,
"name": event.step.name,
"arguments": ""
}
if hasattr(event.step, "arguments") and event.step.arguments:
if isinstance(event.step.arguments, dict):
current_calls[event.index]["arguments"] = json.dumps(event.step.arguments)
else:
current_calls[event.index]["arguments"] = event.step.arguments
elif event.event_type == "step.delta":
if event.delta.type == "arguments":
if event.index in current_calls:
current_calls[event.index]["arguments"] += event.delta.partial_arguments
elif event.delta.type == "text":
print(event.delta.text, end="", flush=True)
elif event.event_type == "interaction.completed":
for index, call in current_calls.items():
args = call["arguments"]
if args:
args = json.loads(args)
else:
args = {}
tool_calls.append({
"type": "function_call",
"id": call["id"],
"name": call["name"],
"arguments": args
})
print(f"\nFinal tool calls ready to execute:")
print(json.dumps(tool_calls, indent=2))
JavaScript
import { GoogleGenAI } from '@google/genai';
const client = new GoogleGenAI({});
const weatherTool = {
type: 'function',
name: 'get_weather',
description: 'Gets the weather for a given location.',
parameters: {
type: 'object',
properties: {
location: { type: 'string', description: 'The city and state' }
},
required: ['location']
}
};
const stream = await client.interactions.create({
model: 'gemini-3.8-flash',
input: 'What is the weather in Paris?',
tools: [weatherTool],
stream: true,
});
const currentCalls = new Map();
let toolCalls = [];
for await (const event of stream) {
const evType = event.event_type;
if (evType === 'step.start') {
if (event.step.type === 'function_call') {
currentCalls.set(event.index, {
id: event.step.id,
name: event.step.name,
arguments: ''
});
if (event.step.arguments) {
if (typeof event.step.arguments === 'object') {
currentCalls.get(event.index).arguments = JSON.stringify(event.step.arguments);
} else {
currentCalls.get(event.index).arguments = event.step.arguments;
}
}
}
} else if (evType === 'step.delta') {
if (event.delta.type === 'arguments') {
if (currentCalls.has(event.index)) {
currentCalls.get(event.index).arguments += event.delta.partial_arguments;
}
} else if (event.delta.type === 'text') {
process.stdout.write(event.delta.text);
}
} else if (evType === 'interaction.completed' || evType === 'interaction.complete') {
toolCalls = Array.from(currentCalls.values()).map(call => ({
type: 'function_call',
id: call.id,
name: call.name,
arguments: call.arguments ? JSON.parse(call.arguments) : {}
}));
console.log('\nFinal tool calls ready to execute:');
console.log(JSON.stringify(toolCalls, null, 2));
}
}
Java
import com.google.genai.Client;
import com.google.genai.gaos.models.interactions.CreateModelInteraction;
import com.google.genai.gaos.models.interactions.Function;
import com.google.genai.gaos.models.interactions.FunctionCallStep;
import com.google.genai.gaos.models.interactions.Interaction;
import com.google.genai.gaos.models.interactions.InteractionsInput;
import com.google.genai.gaos.models.interactions.Model;
import com.google.genai.gaos.models.interactions.Step;
import com.google.genai.gaos.models.operations.CreateInteractionRequestBody;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.Map;
Client client = new Client();
Map<String, Object> parameters = new HashMap<>();
parameters.put("type", "object");
Function function = Function.builder()
.name("custom_function")
.description("A custom function.")
.parameters(parameters)
.build();
CreateModelInteraction params = CreateModelInteraction.builder()
.model(Model.of("gemini-3.6-flash"))
.input(InteractionsInput.of("Call the function."))
.tools(Arrays.asList(function))
.build();
Interaction interaction =
client.interactions.create(CreateInteractionRequestBody.of(params)).interaction().get();
if (interaction.steps().isPresent()) {
for (Step step : interaction.steps().get()) {
if (step instanceof FunctionCallStep) {
FunctionCallStep fc = (FunctionCallStep) step;
System.out.println("Function: " + fc.name().orElse(""));
}
}
}
Go
package main
import (
"context"
"fmt"
"log"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
getWeather := &genai.FunctionDeclaration{
Name: "get_weather",
Description: "Gets the weather for a given location.",
Parameters: &genai.Schema{
Type: genai.TypeObject,
Properties: map[string]*genai.Schema{
"location": {
Type: genai.TypeString,
Description: "The city and state",
},
},
Required: []string{"location"},
},
}
config := &genai.GenerateContentConfig{
Tools: []*genai.Tool{
{FunctionDeclarations: []*genai.FunctionDeclaration{getWeather}},
},
}
for resp, err := range client.Models.GenerateContentStream(
ctx,
"gemini-3.8-flash",
genai.Text("What is the weather in Paris?"),
config,
) {
if err != nil {
log.Fatal(err)
}
for _, fc := range resp.FunctionCalls() {
fmt.Printf("Function to call: %s\n", fc.Name)
fmt.Printf("Arguments: %v\n", fc.Args)
}
}
}
REST
curl -X POST "https://generativelanguage.googleapis.com/v1beta/interactions?alt=sse" \
-H "Content-Type: application/json" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-d '{
"model": "gemini-3.8-flash",
"input": "What is the weather in Paris?",
"tools": [{
"type": "function",
"name": "get_weather",
"description": "Gets the weather for a given location.",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state"}
},
"required": ["location"]
}
}],
"stream": true
}'
Bonnes pratiques
- Descriptions des fonctions et des paramètres : soyez clair et précis.
- Nommage : utilisez des noms descriptifs sans espaces ni caractères spéciaux.
- Typage fort : utilisez des types spécifiques (entier, chaîne, énumération).
- Sélection d'outils : définissez le nombre d'outils actifs sur 10 à 20 maximum.
- Prompt engineering : fournissez du contexte et des instructions.
- Validation : validez les appels de fonction avant de les exécuter.
- Gestion des erreurs : mettez en œuvre une gestion robuste des erreurs.
- Sécurité : utilisez l'authentification appropriée pour les API externes.
Solutions de contournement pour les exigences relatives au texte avant l'outil
Problème : votre requête exige que le modèle génère du texte structuré (XML, YAML, JSON, etc.). (par exemple, <UPDATE>...</UPDATE>) immédiatement avant d'effectuer un appel d'outil, l'appel d'outil peut parfois échouer et générer une erreur Malformed_Function_Call.
Solutions : les solutions de contournement suivantes permettent de résoudre ce problème :
- RECOMMANDÉ : Demandez au modèle de placer ses notes avant l'outil dans un appel de fonction
update()dédié au lieu d'un texte brut (voir ci-dessous). - Demandez au modèle d'écrire des notes sous forme d'en-têtes Markdown (
# UPDATE,## PLAN) au lieu de texte structuré. - Ne demandez pas au modèle de générer du texte avant les appels d'outils.
Solution de contournement recommandée : encapsuler les notes de travail dans un appel de fonction dédié
Au lieu de l'instruction d'origine :
Before calling a tool, in every response you MUST first output a single `<UPDATE>` part as specified, don't skip this part or any of required sub-tags within `<UPDATE>`.
Utilisez cette nouvelle instruction :
Before calling any other tool, in every response you MUST first call `update` with all required parameters (previous_step, plan, next_step, external).
Mettez également à jour toutes les références à l'ancien format XML <UPDATE> dans la demande du client. Ajoutez ensuite la déclaration de fonction correspondante pour la fonction de mise à jour :
{
"name": "update",
"description": "Update working notes (previous step analysis, plan, next step, external note).",
"parameters": {
"type": "OBJECT",
"properties": {
"previous_step": {
"type": "STRING",
"description": "Key findings and outcomes since the previous step."
},
"plan": {
"type": "STRING",
"description": "The current status of the plan."
},
"next_step": {
"type": "STRING",
"description": "Brief explanation of the immediate next action according to the plan."
},
"external": {
"type": "STRING",
"description": "A short, plain-language note shown to the User about what you are ABOUT TO DO next."
}
},
"required": [
"previous_step",
"plan",
"next_step",
"external"
]
}
}
Le modèle effectue ensuite deux appels au cours de la même étape : l'appel update() qui remplace le code XML structuré et l'appel de fonction qu'il souhaite effectuer.
Remarques et limites
- Seul un sous-ensemble du schéma OpenAPI est accepté.
- En mode
any, l'API peut refuser les schémas très volumineux ou profondément imbriqués. - Les types de paramètres acceptés dans Python sont limités.