Updated daily at 05:00 UTC
Benchmark
Agent Arena (arena.ai)
Measures: Ranks models by how well they run real users' agentic sessions (code, chat, work), using signals such as confirmed task success, praise vs complaint, steerability, bash recovery and tool hallucination; the headline is Net Improvement (%) with a +/- interval.
- Maintainer
- Arena Intelligence (arena.ai; formerly LMArena / LMSYS Chatbot Arena)
- Introduced
- —
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Independent
- In scoring set
- Yes
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results