Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
docs(gemini): update to use gemini 3.7 and new models
Mark McDonald committed
Aug 25, 2026 at 15:05 UTC
e3d2c093e39463861a07650a9f903070eaa6243c
4 files changed
+9
-8
python/providers/gemini/gemini_demo.py
+1
-1
@@ -21,6 +21,6 @@ config = types.GenerateContentConfig(
21
)
22
23
# Use the chat interface.
24
-chat = client.chats.create(model="gemini-2.0-flash", config=config)
24
+chat = client.chats.create(model="gemini-3.7-flash", config=config)
25
response = chat.send_message("Can you star composiohq/composio repository on github")
26
print(response.text)
python/providers/google/README.md
+1
-1
@@ -30,7 +30,7 @@ session = composio.create(user_id="user_123")
30
tools = session.tools()
31
32
model = GenerativeModel(
33
- "gemini-2.0-flash",
33
+ "gemini-3.7-flash",
34
tools=[Tool(function_declarations=tools)],
35
)
36
chat = model.start_chat()
python/providers/google/google_demo.py
+1
-1
@@ -18,7 +18,7 @@ composio = Composio(provider=GoogleProvider())
18
tool = composio.tools.get(user_id="default", toolkits=["GITHUB"])
19
20
# Initialize the Gemini model
21
-model = GenerativeModel("gemini-1.5-pro", tools=[tool])
21
+model = GenerativeModel("gemini-3.7-flash", tools=[tool])
22
23
# Start a chat session
24
chat = model.start_chat()
ts/docs/providers/google.md
+6
-5
@@ -41,7 +41,7 @@ const task = "Fetch the details of the user 'haxzie'";
41
42
// Generate content with function calling
43
const response = await ai.models.generateContent({
44
- model: 'gemini-2.0-flash-001',
44
+ model: 'gemini-3.7-flash',
45
contents: task,
46
config: {
47
tools: [{ functionDeclarations: tools }],
@@ -88,10 +88,11 @@ const ai = new GoogleGenAI({
88
89
## Supported Models
90
91
-- `gemini-2.0-flash-001` - Fast, efficient model for most use cases
92
-- `gemini-2.0-pro-001` - Advanced model with enhanced capabilities
93
-- `gemini-1.5-flash-001` - Legacy model for backward compatibility
94
-- `gemini-1.5-pro-001` - Legacy model with advanced capabilities
91
+- `gemini-3.7-flash` - Fast, efficient default model for most use cases
92
+- `gemini-3.1-pro-preview` - Advanced model with enhanced reasoning and coding capabilities
93
+- `gemini-3.5-flash-lite` - Fastest and lowest-cost option for lightweight tasks
94
+- `gemini-2.5-flash` - Fast model with thinking capabilities
95
+- `gemini-2.5-pro` - Advanced reasoning model
96
97
## Function Calling
98