What it does
The server lets an MCP client run benchmarks of AI models against real prompts. According to the publisher, it covers more than 340 models and is meant to help identify cheaper and faster alternatives to a model currently in use.
It would suit developers who want to compare model cost and speed on their own prompts before changing models in an application or workflow. The publisher's description does not detail the specific comparison workflow or output format.
The source gives no setup steps. Connecting it to an MCP client is likely required, but the publisher's facts do not specify the configuration.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.tjacquesson/llmtest-mcp
Benchmark AI models on real prompts. Find cheaper, faster alternatives across 340+ models.
v0.6.1As shown on registry.modelcontextprotocol.io when we last read it