- Context
- 1Mtokens
- Released
- 1y ago
Benchmarks
DeepSWE is the primary coding benchmark where an exact model evaluation exists. Arena leaderboards and other sourced evaluations appear below; missing scores are not zero.
How it ranks
Against its predecessor, similarly priced models and the models ranked around it.
DeepSWE 1.1
Best pass@1 per model · Higher is better
Not evaluated in DeepSWE 1.1
Arena Text
Best Arena rating per model · Higher is better
Arena #197
- 1426
Claude Opus 4Similar release date - 1422
Qwen3 235B A22B Instruct 2507Similar release date - 1421
R1 0528Similar release date - 1379
Trinity Large PreviewNearby rank - 1378
GLM 4.6VNearby rank - 1375
Gemini 2.5 Flash-Lite Preview 06-17 - 1374Claude Sonnet 3.5 v2Nearby rank
- 1373GLM 4.5 AirNearby rank
- 1356
Mistral Small 3.2Similar release date
Ahead of 1 of 4 models released around the same time
Arena WebDev
Best Arena rating per model · Higher is better
Not ranked on Arena WebDev yet
Arena Vision
Best Arena rating per model · Higher is better
Arena #100
- 1211
GPT-5Similar release date - 1208Claude Opus 4Similar release date
- 1196
Grok 4.1 Fast (Reasoning)Nearby rank - 1193Claude Sonnet 3.7Nearby rank
- 1189Gemini 2.5 Flash-Lite Preview 06-17
- 1188
Qwen3 VL 235B A22B ThinkingNearby rank - 1185GPT-5 MiniNearby rank
- 1153GLM 4.5VSimilar release date
- 1140Mistral Small 3.2Similar release date
Ahead of 2 of 4 models released around the same time
Arena Search
Best Arena rating per model · Higher is better
Not ranked on Arena Search yet
Arena Document
Best Arena rating per model · Higher is better
Not ranked on Arena Document yet
Agent Arena
Best IPS score per model · Higher is better
Not ranked on Agent Arena yet
Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.
All published scores
- Arena TextArena #1971375
- default1375thinking95% CI 1370 to 1379 · 32,330 votes
- Arena VisionArena #1001189
- default1189thinking95% CI 1181 to 1196 · 35,868 votes
Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.