Skip to content

docs: document You.com as alternative browser backend for gpt-oss#603

Open
tyler5673 wants to merge 2 commits into
vllm-project:mainfrom
tyler5673:docs/gpt-oss-youcom-browser-backend
Open

docs: document You.com as alternative browser backend for gpt-oss#603
tyler5673 wants to merge 2 commits into
vllm-project:mainfrom
tyler5673:docs/gpt-oss-youcom-browser-backend

Conversation

@tyler5673

Copy link
Copy Markdown

The gpt-oss reference library ships a YouComBackend and the MCP browser server (browser_server.py) defaults to BROWSER_BACKEND=youcom, but the vLLM gpt-oss recipes only documented the Exa backend.

This adds a brief "Browser backend" section to both the 20b and 120b recipe guides showing how to use You.com via the MCP browser server with YDC_API_KEY.

The demo tool server path (--tool-server demo) still uses Exa by default and is noted as such. The You.com path applies to the MCP browser server, which defaults to BROWSER_BACKEND=youcom.

The mcp run and vllm serve commands are in separate code blocks since both are long-running blocking processes that need separate terminals.

The gpt-oss reference library ships a YouComBackend and the MCP browser
server (browser_server.py) defaults to BROWSER_BACKEND=youcom, but the
vLLM gpt-oss recipes only documented the Exa backend.

Add a brief 'Browser backend' section to both the 20b and 120b recipe
guides showing how to use You.com via the MCP browser server with
YDC_API_KEY. The mcp run and vllm serve commands are in separate code
blocks since both are long-running blocking processes that need
separate terminals.

Signed-off-by: Tyler Eastman <tyler@you.com>
@vercel

vercel Bot commented Jun 30, 2026

Copy link
Copy Markdown
Contributor

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
vllm-recipes Ready Ready Preview, Comment Jun 30, 2026 7:52pm

Request Review

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Code Review

This pull request updates the documentation for the gpt-oss-120b and gpt-oss-20b models by adding a 'Browser backend' section to guide users on configuring web search using Exa or You.com. The reviewer correctly identified a potential port conflict when running both the MCP browser server and the vLLM server on the same machine, as both default to port 8000. To resolve this, it is recommended to explicitly set a different port (such as PORT=8001) for the MCP server.

Comment thread models/openai/gpt-oss-120b.yaml
Comment thread models/openai/gpt-oss-20b.yaml
The MCP browser server (browser_server.py) defaults to port 8001, not
8000. Replace the vague <browser-server-ip>:<port> placeholder with a
concrete localhost:8001 example and note the default port.

Signed-off-by: Tyler Eastman <tyler@you.com>
@tyler5673 tyler5673 marked this pull request as ready for review June 30, 2026 20:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant