- input
- $0.21
- cached
- $0.042
- output
- $0.84
- Context
- 197Ktokens
- Released
- 7mo ago
- Size
- 229Bparameters
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 #144
- 1483
GLM 5.3Similar price - 1417
DeepSeek V3.1Similar price - 1416
Qwen3.5-122B-A10BNearby rank - 1416DeepSeek V3.1 TerminusNearby rank
- 1415
MiniMax M2.7 - 1415
Mistral Large 3 2512Nearby rank - 1415
GPT-4.1Nearby rank - 1391MiniMax M2.5Predecessor
- 1303Qwen2.5 72B InstructSimilar price
- 1303Mistral Small 3.1 24BSimilar price
+24 vs MiniMax M2.5 (predecessor) · Ahead of 2 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Arena #106
- 1622GLM 5.3Similar price
- 1519GPT-5.6 LunaSimilar price
- 1400Qwen3.5 397B A17BNearby rank
- 1399GLM 5V TurboNearby rank
- 1398MiniMax M2.7
- 1398GPT-5.4 MiniNearby rank
- 1397GPT-5.1Nearby rank
- 1386MiniMax M2.5Predecessor
- 1270DeepSeek V3.2 ExpSimilar price
- 1253Qwen3.5-35B-A3BSimilar price
Level with MiniMax M2.5 (predecessor, overlapping 95% CIs) · Ahead of 2 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
Terminal Bench
Best score per model · Higher is better
#23 of 49
- 61.6%GPT-5.1-Codex-MiniSimilar price
- 47.6%GPT-5.1Nearby rank
- 46.5%
Claude Sonnet 4.5Nearby rank - 45.1%MiniMax M2.7
- 44.3%GPT-5-CodexNearby rank
- 43.2%
Kimi K2.5Nearby rank - 42.7%MiniMax M2.5Predecessor
- 39.6%DeepSeek V3.2Similar price
- 24.6%Qwen3.6 35B A3BSimilar price
- 18.7%gpt-oss-120bSimilar price
+2.4 pts vs MiniMax M2.5 (predecessor) · Ahead of 3 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 #1441415
+24 vs MiniMax M2.5 (predecessor)
- default141595% CI 1412 to 1418 · 89,504 votes
- Arena WebDevArena #1061398
Level with MiniMax M2.5 (predecessor, overlapping 95% CIs)
- default139895% CI 1392 to 1404 · 15,728 votes
- Terminal Bench#23 of 4945.1%
+2.4 pts vs MiniMax M2.5 (predecessor)
- default45.1%IndusAGI Coding Agent
- default42.9%Harness Agent
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.