- input
- $0.60
- cached
- $0.11
- output
- $2.20
- Context
- 131Ktokens
- Released
- 1y ago
- Size
- 358Bparameters
Cost/1M tokens
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 #154
- 1450
Kimi K2.5Similar price - 1426
Nemotron 3 UltraSimilar price - 1423
Mistral Large 4Similar price - 1421
R1 0528Similar price - 1413
Qwen3 VL 235B A22B InstructNearby rank - 1412
Inkling 2 SmallNearby rank - 1411
GLM 4.5 - 1409
Gemini 2.5 FlashNearby rank - 1409
Hy3 previewNearby rank
Ahead of 0 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Not ranked on Arena WebDev yet
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
Humanity's Last Exam
Best score per model · Higher is better
#23 of 34
- 44.5%Gemini 3.8 FlashSimilar price
- 24.4%Kimi K2.5Similar price
- 19.4%
GPT-5 MiniSimilar price - 11.0%Gemini 2.5 Flash Preview 05-20Nearby rank
- 8.6%Gemini 3.1 Flash LiteNearby rank
- 8.3%GLM 4.5
- 8.1%GLM 4.5 AirNearby rank
- 8.0%
Claude Sonnet 3.7Nearby rank - 4.5%Mistral Medium 3Similar price
Ahead of 1 of 4 similarly priced models
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 #1541411
- default141195% CI 1406 to 1416 · 23,759 votes
- Humanity's Last Exam#23 of 348.3%
- default8.3%
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.