- Context
- 205Ktokens
- Released
- 1y ago
- Size
- 229Bparameters
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 #232
- 1486
Gemini 3 Pro PreviewSimilar release date - 1455
GPT-5.1Similar release date - 1430
Kimi K2 Thinking TurboSimilar release date - 1414
Claude Haiku 4.5Similar release date - 1346GPT-4o (2024-05-13)Nearby rank
- 1346
Qwen-PlusNearby rank - 1344
MiniMax-M2 - 1343
Llama 3.3 Nemotron Super 49B v1.5Nearby rank - 1342Gemma 3 12BNearby rank
Ahead of 0 of 4 models released around the same time
Arena WebDev
Best Arena rating per model · Higher is better
Arena #136
- 1441Gemini 3 Pro PreviewSimilar release date
- 1397GPT-5.1Similar release date
- 1330Claude Haiku 4.5Similar release date
- 1322Kimi K2 Thinking TurboSimilar release date
- 1303
Laguna XS.2Nearby rank - 1299MiniMax-M2
- 1270
DeepSeek V3.2 ExpNearby rank - 1263
Mistral Medium 3.5Nearby rank
Ahead of 0 of 4 models released around the same time
Arena Search
Best Arena rating per model · Higher is better
Not ranked on Arena Search yet
Agent Arena
Best IPS score per model · Higher is better
Not ranked on Agent Arena yet
Terminal Bench
Best score per model · Higher is better
#38 of 49
- 69.4%Gemini 3 Pro PreviewSimilar release date
- 61.6%GPT-5.1-Codex-MiniSimilar release date
- 35.7%Kimi K2 ThinkingSimilar release date
- 35.5%Claude Haiku 4.5Similar release date
- 33.4%
GLM 4.7Nearby rank - 32.6%Gemini 2.5 ProNearby rank
- 30.0%MiniMax-M2
- 27.8%Kimi K2 InstructNearby rank
- 25.8%
x-AI/Grok-Code-Fast 1Nearby rank
Ahead of 0 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 #2321344
- default134495% CI 1336 to 1351 · 6,832 votes
- Arena WebDevArena #1361299
- default129995% CI 1288 to 1310 · 6,599 votes
- Terminal Bench#38 of 4930.0%
- default30.0%Terminus 2
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.