Ran it, or it didn’t happen.

Benchmark open models on your own hardware. Runs count once their event log checks out.

Run your first benchmark
Open models
250Open models

+36 in 12 weeks

Closed models
191Closed models
Labs
57Labs
Runnable with BenchClub
38Runnable with BenchClub

Open-weight models

Score by size

Plot against
Best for its size
Gap
Coding
−55
Math
−41
Expert
−64
Chinese
−96
140015001600

Arena scores: LMArena, CC BY 4.0, 2026-11-11. Measured on served versions.

Discussed, last 30 days

On BenchClub