- input
- $1.20
- cached
- $0.24
- output
- $4.00
- Context
- 203Ktokens
- Released
- 7mo ago
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 #110
- 1496
Muse Spark 1.3Similar price - 1447
GPT-5.4 MiniSimilar price - 1443
InklingSimilar price - 1435
Beam 1.0 501B A23BNearby rank - 1434
Qwen3 Max PreviewNearby rank - 1434
GLM 5V Turbo - 1434
MiMo-V2.5Nearby rank - 1433
Qwen3.8 27BNearby rank - 1414
Claude Haiku 4.5Similar price
Ahead of 1 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Arena #105
- 1679GPT-6.1 TerraSimilar price
- 1655Muse Spark 1.3Similar price
- 1414InklingSimilar price
- 1406GPT-5.3-CodexNearby rank
- 1400Qwen3.5 397B A17BNearby rank
- 1399GLM 5V Turbo
- 1398
MiniMax M2.7Nearby rank - 1398GPT-5.4 MiniNearby rank
- 1330Claude Haiku 4.5Similar price
Ahead of 1 of 4 similarly priced models
Arena Vision
Best Arena rating per model · Higher is better
Arena #72
- 1294Muse Spark 1.2Similar price
- 1293
Gemini 3.8 FlashSimilar price - 1250GPT-5.4 MiniSimilar price
- 1242
Grok 4.3Similar price - 1234MiMo-V2.5Nearby rank
- 1232MiniMax M3Nearby rank
- 1232GLM 5V Turbo
- 1231
Mistral Large 4Nearby rank - 1226Qwen3.5-122B-A10BNearby rank
Ahead of 0 of 4 similarly priced models
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
Arena #46
- 1480Muse Spark 1.1Similar price
- 1469Gemini 3.8 FlashSimilar price
- 1450
Kimi K2.6Similar price - 1419Claude Haiku 4.5Similar price
- 1416GPT-5.1Nearby rank
- 1406GLM 5V Turbo
Ahead of 0 of 4 similarly priced models
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 #1101434
- default143495% CI 1428 to 1441 · 10,291 votes
- Arena WebDevArena #1051399
- default139995% CI 1386 to 1412 · 2,375 votes
- Arena VisionArena #721232
- default123295% CI 1226 to 1237 · 43,637 votes
- Arena DocumentArena #461406
- default140695% CI 1398 to 1413 · 9,136 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.