- Context
- 200Ktokens
- Released
- 2y 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 #199
- 1378
GLM 4.6VNearby rank - 1375
Gemini 2.5 Flash-Lite Preview 06-17Nearby rank - 1374
Claude Sonnet 3.5 v2 - 1373GLM 4.5 AirNearby rank
- 1369
Qwen3-Next 80B-A3B (Thinking)Nearby rank - 1271
Qwen2.5 Coder 32B InstructSimilar release date - 1267
Aya Expanse 32BSimilar release date - 1262Claude Haiku 3Predecessor
- 1238
Ministral 8B InstructSimilar release date - 1167
Llama 3.2 3B InstructSimilar release date
+113 vs Claude Haiku 3 (predecessor) · Ahead of 4 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 #111
- 1172Gemini 2.0 FlashNearby rank
- 1162
GPT-4o (2024-05-13)Nearby rank - 1161Claude Sonnet 3.5 v2
- 1161GLM 4.6VNearby rank
- 1160Gemma 3 27BNearby rank
- 1045
Molmo 72BSimilar release date - 1032Llama-3.2-90B-Vision-InstructSimilar release date
- 1025Pixtral 12BSimilar release date
- 1019
Nova Lite 1.0Similar release date - 1000Claude Haiku 3Predecessor
+161 vs Claude Haiku 3 (predecessor) · Ahead of 4 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
FrontierMath-2025-02-28-Private
Best score per model · Higher is better
#48 of 58
- 3.8%GLM 4.6Nearby rank
- 2.4%GLM 4.7Nearby rank
- 2.1%Claude Sonnet 3.5 v2
- 1.7%
DeepSeek V3Nearby rank - 1.0%Qwen MaxNearby rank
- 0.3%Mistral Large 2.1Similar release date
- 0.3%GPT-4o (2024-08-06)Similar release date
- 0.3%GPT-4o (2024-11-20)Similar release date
- 0.0%Llama 4 ScoutSimilar release date
Ahead of 4 of 4 models released around the same time
FrontierMath-Tier-4-2025-07-01-Private
Best score per model · Higher is better
#39 of 46
- 2.1%o3Similar release date
- 2.1%GPT-5.4 MiniNearby rank
- 0.0%Claude Sonnet 3.5 v2
- 0.0%Claude Sonnet 4Nearby rank
- 0.0%
Kimi K2 ThinkingNearby rank - 0.0%GPT-4.1Similar release date
Ahead of 0 of 2 models released around the same time
GPQA diamond
Best score per model · Higher is better
#131 of 161
- 56.1%Magistral SmallNearby rank
- 55.6%Qwen3 30B A3B Instruct 2507Nearby rank
- 55.3%Claude Sonnet 3.5 v2
- 51.8%Llama 4 ScoutNearby rank
- 51.3%Mistral Large 2.1Similar release date
- 47.9%GPT-4o (2024-11-20)Similar release date
- 41.8%Qwen TurboSimilar release date
- 41.0%Llama-3.2-90B-Vision-InstructSimilar release date
- 36.3%Claude Haiku 3Predecessor
+19.0 pts vs Claude Haiku 3 (predecessor) · Ahead of 4 of 4 models released around the same time
Humanity's Last Exam
Best score per model · Higher is better
#33 of 34
- 20.3%o3Similar release date
- 5.7%Llama 4 MaverickSimilar release date
- 5.4%GPT-4.1Similar release date
- 4.5%Mistral Medium 3Nearby rank
- 4.1%Claude Sonnet 3.5 v2
- 2.7%GPT-4o (2024-11-20)Similar release date
Ahead of 1 of 4 models released around the same time
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#136 of 158
- 16.1%Qwen MaxNearby rank
- 13.8%
Phi 4Nearby rank - 8.5%Claude Sonnet 3.5 v2
- 8.5%Mistral Large 2407Nearby rank
- 8.1%Qwen2.5 72B InstructNearby rank
- 7.8%Mistral Large 2.1Similar release date
- 6.3%GPT-4o (2024-11-20)Similar release date
- 6.1%Qwen TurboSimilar release date
- 2.6%Llama-3.2-90B-Vision-InstructSimilar release date
- 1.8%Claude Haiku 3Predecessor
+6.7 pts vs Claude Haiku 3 (predecessor) · Ahead of 4 of 4 models released around the same time
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 #1991374
+113 vs Claude Haiku 3 (predecessor)
- default137495% CI 1371 to 1378 · 87,743 votes
- Arena VisionArena #1111161
+161 vs Claude Haiku 3 (predecessor)
- default116195% CI 1153 to 1169 · 10,372 votes
- FrontierMath-2025-02-28-Private#48 of 582.1%
- default2.1%
Source:Epoch AI evaluation - FrontierMath-Tier-4-2025-07-01-Private#39 of 460.0%
- default0.0%
Source:Epoch AI evaluation - GPQA diamond#131 of 16155.3%
+19.0 pts vs Claude Haiku 3 (predecessor)
- default55.3%
Source:Epoch AI evaluation - Humanity's Last Exam#33 of 344.1%
- default4.1%
- OTIS Mock AIME 2024-2025#136 of 1588.5%
+6.7 pts vs Claude Haiku 3 (predecessor)
- default8.5%
Source:Epoch AI evaluation
Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.
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.