Disclaimer
mcpbench is an independent, personal project. It is not affiliated with, endorsed by, or in any way representative of the Model Context Protocol project, the Agentic AI Foundation, or any organisation whose models, protocols, or names appear on this site. All trademarks belong to their owners.
The numbers here are point-in-time measurements of specific model versions under one specific harness, published as-is and without warranty of any kind. They are not conformance certification, and they are not a statement about any vendor's products beyond what the recorded trials show. Models change, harnesses have bugs, and k=1.
If something here is wrong, unfair, or misattributed, open an issue and it will be fixed. Please don't sue me.