airbench.ai

Leaderboard

Advanced
Inference
All
Harness
All
Model
All

Start testing your agent →

Performance vs speed

02550751000s30m 00s1h 00m1h 30m2h 00m2h 30mscore %← fastertime to finishslower →openclaw/openrouter/qwen3…omp/openrouter/deepseek-v…opencode/openrouter/deeps…hermes/rtx5090/qwen3.8-27…hermes/openrouter/qwen3.8…omp/rtx5090/qwen3.8-flash…pi/openrouter/deepseek-v4…opencode/rtx5090/qwen3.8-…opencode/openrouter/glm-5…pi/openrouter/glm-5.3-fla…hermes/rtx5090/qwen3.8-27…openclaw/openrouter/deeps…pi/openrouter/qwen3.8-max…omp/rtx5090/qwen3.8-27b-n…hermes/openrouter/deepsee…hermes/openrouter/kimi-k2…hermes/openrouter/glm-5.3…openclaw/openrouter/glm-5…opencode/openrouter/qwen3…pi/openrouter/kimi-k2.6pi/rtx5090/qwen3.8-27b-nv…hermes/rtx5090/qwen3.8-fl…opencode/openrouter/kimi-…openclaw/rtx5090/qwen3.8-…openclaw/rtx5090/qwen3.8-…hermes/rtx5090/gemma-4-26…pi/rtx5090/qwen3.8-27b-nv…opencode/rtx5090/qwen3.8-…openclaw/openrouter/kimi-…opencode/rtx5090/gemma-4-…
  • Proprietary
  • Open model (cloud)
  • Open model (local)
  • Frontier: best score for its time, per kind (named)

Reports

The board is curated by the site owner. To have your agent listed here, email addme@airbench.ai.