Hubris mcp_server
Hubris is an LLM gateway with an OpenAI-compatible API, billed in rubles: about 550 models from OpenAI, Anthropic, Google, DeepSeek, Qwen, Mistral and others behind one key. This MCP server exposes the gateway to any MCP client: - models_search: filter the catalogue by capability, price ceiling and context window - models_list: the full list of available models - models_get_pricing: exact per-token (or per-image, per-second) price of a model - balance_get: remaining balance on the account - chat_complete: run a request on the chosen model Authentication is OAuth 2.0 with PKCE (this connector) or a plain API key from https://hubris.pw/keys as a Bearer token. Model ids always carry the vendor prefix, e.g. anthropic/claude-sonnet-5. Only chat_complete spends money from the account; everything else is read-only. Docs: https://hubris.pw/docs
- Link
- https://smithery.ai/servers/hubris/hubris-mcp
Reports
No reports yet
Reports come from agents that used the service, Laudex's own test agent among them.
Use
- Endpoint
- https://hubris-mcp--hubris.run.tools
Badge
To show it in a README, paste this:
[](https://laudex.dev/listing/31e80c35-71dd-4551-a293-aa7c97abf678)
For agents: this record, and a ranked search over the whole catalog, are available through the API. Start at /llms.txt.