Skip to content

Support for new Gemini Computer Use Models #1110

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 13 commits into from
Oct 7, 2025
Merged

Support for new Gemini Computer Use Models #1110

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
3ded78e
cheeta updates 1
miguelg719 Oct 7, 2025
e42d3aa
paramaterize mouse cursor
tkattkat Oct 7, 2025
4f6d96c
add more agent logs to openai cua client
Kylejeong2 Oct 7, 2025
2fd73ed
fix env type
sameelarif Oct 7, 2025
451bfb8
update missing api key error message
sameelarif Oct 7, 2025
664b0f3
update model name
sameelarif Oct 7, 2025
b5e1749
updates
miguelg719 Oct 7, 2025
4204000
full text response in anthropic cua client
Kylejeong2 Oct 7, 2025
e3ee7dd
Update cua-example.ts
sameelarif Oct 7, 2025
02f89a0
Update stagehand.config.ts
sameelarif Oct 7, 2025
a68f69c
changeset
miguelg719 Oct 7, 2025
4b519fa
Update prompt.ts
sameelarif Oct 7, 2025
b60bd28
Merge branch 'cheetah' of https://github.com/browserbase/stagehand in…
sameelarif Oct 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .changeset/wicked-ducks-share.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
---
"@browserbasehq/stagehand": patch
---

Add support for new Gemini Computer Use models
9 changes: 5 additions & 4 deletions examples/cua-example.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,14 +26,15 @@ async function main() {

// Create a computer use agent
const agent = stagehand.agent({
provider: "openai",
// For Anthropic, use claude-sonnet-4-20250514 or claude-3-7-sonnet-latest
model: "computer-use-preview",
provider: "google",
// For Anthropic, use claude-sonnet-4-20250514 or claude-sonnet-4-5-20250929
// For OpenAI, use computer-use-preview-03-11
model: "gemini-2.5-computer-use-preview-10-2025",
instructions: `You are a helpful assistant that can use a web browser.
You are currently on the following page: ${page.url()}.
Do not ask follow up questions, the user will trust your judgement.`,
options: {
apiKey: process.env.OPENAI_API_KEY,
apiKey: process.env.GOOGLE_API_KEY,
},
});

Expand Down
11 changes: 10 additions & 1 deletion lib/agent/AgentProvider.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@ import { ToolSet } from "ai/dist";
import { AgentClient } from "./AgentClient";
import { AnthropicCUAClient } from "./AnthropicCUAClient";
import { OpenAICUAClient } from "./OpenAICUAClient";
import { GoogleCUAClient } from "./GoogleCUAClient";

// Map model names to their provider types
export const modelToAgentProviderMap: Record<string, AgentType> = {
Expand All @@ -16,6 +17,7 @@ export const modelToAgentProviderMap: Record<string, AgentType> = {
"claude-3-7-sonnet-latest": "anthropic",
"claude-sonnet-4-20250514": "anthropic",
"claude-sonnet-4-5-20250929": "anthropic",
"gemini-2.5-computer-use-preview-10-2025": "google",
};

/**
Expand Down Expand Up @@ -64,9 +66,16 @@ export class AgentProvider {
clientOptions,
tools,
);
case "google":
return new GoogleCUAClient(
type,
modelName,
userProvidedInstructions,
clientOptions,
);
Copy link
Contributor

@greptile-apps greptile-apps bot Oct 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

logic: GoogleCUAClient constructor is missing the tools parameter that OpenAI and Anthropic clients receive

Prompt To Fix With AI 
This is a comment left during a code review.
Path: lib/agent/AgentProvider.ts
Line: 75:75

Comment:
**logic:** GoogleCUAClient constructor is missing the `tools` parameter that OpenAI and Anthropic clients receive

How can I resolve this? If you propose a fix, please make it concise.

default:
throw new UnsupportedModelProviderError(
["openai", "anthropic"],
["openai", "anthropic", "google"],
"Computer Use Agent",
);
}
Expand Down
4 changes: 2 additions & 2 deletions lib/agent/AnthropicCUAClient.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@ export class AnthropicCUAClient extends AgentClient {
private baseURL?: string;
private client: Anthropic;
public lastMessageId?: string;
private currentViewport = { width: 1024, height: 768 };
private currentViewport = { width: 1288, height: 711 };
private currentUrl?: string;
private screenshotProvider?: () => Promise<string>;
private actionHandler?: (action: AgentAction) => Promise<void>;
Expand Down Expand Up @@ -290,7 +290,7 @@ export class AnthropicCUAClient extends AgentClient {

logger({
category: "agent",
message: `Found text block: ${textBlock.text.substring(0, 50)}...`,
message: `Found text block: ${textBlock.text}`,
level: 2,
});
} else {
Expand Down
Loading