- input
- $0.14
- cached
- $0.05
- output
- $0.80
- Context
- 262Ktokens
- Released
- 1y ago
- Size
- 235Bparameters
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 #132
- 1425
DeepSeek V3.2 ExpSimilar price - 1423
Mistral Large 4Nearby rank - 1422
Claude Haiku 5.5Nearby rank - 1422
Qwen3 235B A22B Instruct 2507 - 1421R1 0528Nearby rank
- 1418
X-Ai/Grok-4-Fast-Non-ReasoningNearby rank - 1409
Hy3 previewSimilar price - 1387
Solar Pro 4Similar price - 1352
gpt-oss-120bSimilar price
Ahead of 3 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
ARC-AGI-2
Best score per model · Higher is better
#73 of 85
- 61.3%Claude Haiku 5.5Similar price
- 4.9%
MiniMax M2.5Similar price - 4.0%DeepSeek V3.2Similar price
- 2.5%
Gemini 2.5 Flash Preview 05-20Nearby rank - 1.3%Gemini 3 Flash PreviewNearby rank
- 1.3%Qwen3 235B A22B Instruct 2507
- 1.1%R1 0528Nearby rank
- 0.9%Claude Sonnet 3.7Nearby rank
- 0.0%GPT-4o-mini (2024-07-18)Similar 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 #1321422
- default142295% CI 1420 to 1425 · 97,642 votes
- ARC-AGI-2#73 of 851.3%
- default1.3%
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.