Guides
Web search
Web search that doesn't depend on the model
Native web search is tied to particular providers. OpenLLM exposes search as an MCP tool backed by the search provider you configure, so any model behind your agent can use it.
One search owner per turn
Where a provider has native search, its own tool passes through unchanged. Elsewhere, the agent calls the MCP search tool explicitly — the gateway never runs hidden searches or follow-up turns.
Code search and memory over MCP
The same binary ships one MCP server: semantic search over your indexed repositories and documentation, plus memory that persists across sessions and clients.
Questions
Something else? Contact us