A small benchmark tries to compare Hermes Agent with rivals
is a benchmark meant to compare such as PI, OpenCode, and under the same conditions. The test keeps the model fixed as Qwen 3.5 9B, so the comparison focuses more on the harness and less on the model.
The benchmark is still , and its or practical value is not yet proven. The is public, so others can inspect it, fork it, and change the tests.
Key points
- The comparison includes PI, OpenCode, and .
- The model is kept constant as Qwen 3.5 9B.
- The benchmark code is public and can be edited or forked.
- Its accuracy and real-world usefulness are still uncertain.