- input
- $0.40
- cached
- $0.08
- output
- $2.00
- Context
- 1Mtokens
- Released
- 6mo ago
- Size
- 311Bparameters
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 #111
- 1455
Gemini 3.5 Flash LiteSimilar price - 1434Qwen3 Max PreviewNearby rank
- 1434
GLM 5V TurboNearby rank - 1434
MiMo-V2.5 - 1433
Qwen3.8 27BNearby rank - 1433
Gemini 3.1 Flash Lite PreviewNearby rank - 1409
Qwen3.5-27BSimilar price - 1408
Mistral Medium 3.1Similar price - 1335
Nova 2 LiteSimilar price
Ahead of 3 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Arena #88
- 1446
Gemini 3.1 Pro PreviewNearby rank - 1441
Gemini 3 Pro PreviewNearby rank - 1438
MiMo-V2.5 - 1437
Beam 1.0 501B A23BNearby rank - 1436
Kimi K2.5Nearby rank - 1435
Gemini 3.5 Flash LiteSimilar price - 1357
Qwen3.5-27BSimilar price - 1342
GLM 4.6Similar price - 1196
Devstral 2 2512Similar price
Ahead of 3 of 4 similarly priced models
Arena Vision
Best Arena rating per model · Higher is better
Arena #70
- 1266
Gemini 3.5 Flash LiteSimilar price - 1235
GPT-6 LunaNearby rank - 1235
Gemini 3.1 Flash Lite PreviewNearby rank - 1234
MiMo-V2.5 - 1232MiniMax M3Nearby rank
- 1232
GLM 5V TurboNearby rank - 1216
Qwen3.5-27BSimilar price - 1160
Mistral Medium 3.1Similar price - 1153
GLM 4.5VSimilar price
Ahead of 3 of 4 similarly priced models
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
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 #1111434
- default143495% CI 1429 to 1438 · 46,820 votes
- Arena WebDevArena #881438
- default143895% CI 1432 to 1444 · 14,632 votes
- Arena VisionArena #701234
- default123495% CI 1228 to 1240 · 29,427 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.