- input
- $0.60
- cached
- $0.16
- output
- $3.50
- Context
- 262Ktokens
- Released
- 5mo ago
- Size
- 1Tparameters
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
#26 of 28
- 74.1%
- 73.8%
- 73.6%
- 72.7%
- 69.9%
- 69.6%
- 69.0%
- 68.5%
- 67.5%
- 67.2%
- 30.5%Kimi K2.7 Code
Arena Text
Best Arena rating per model · Higher is better
Not ranked on Arena Text yet
Arena WebDev
Best Arena rating per model · Higher is better
Arena #77
- 1588Gemini 3.7 FlashSimilar price
- 1581
DeepSeek V4 Pro 0813Similar price - 1573
Step 5 PreviewSimilar price - 1482
Qwen3.6 Max PreviewNearby rank - 1479
MiMo-V2.5-ProNearby rank - 1474Kimi K2.7 Code
- 1462GPT-5.4Nearby rank
- 1461
Qwen3.6 PlusNearby rank - 1400Qwen3.5 397B A17BSimilar price
Ahead of 1 of 4 similarly priced models
Arena Vision
Best Arena rating per model · Higher is better
Not ranked on Arena Vision 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
FrontierMath-Tier-4-v2-Private
Best score per model · Higher is better
#57 of 72
- 36.6%Gemini 3.7 FlashSimilar price
- 26.8%DeepSeek V4 Pro 0813Similar price
- 26.8%
Mistral Large 4Similar price - 17.1%GLM 5.3 FlashNearby rank
- 14.6%Grok 4.3Similar price
- 12.2%Kimi K2.7 Code
- 12.2%GPT-5 MiniNearby rank
- 12.2%GPT-5.4 NanoNearby rank
Ahead of 0 of 4 similarly priced models
FrontierMath-Tiers-1-3-v2-Private
Best score per model · Higher is better
#51 of 87
- 71.6%Gemini 3.7 FlashSimilar price
- 64.6%DeepSeek V4 Pro 0813Similar price
- 55.8%GLM 5.3 FlashNearby rank
- 55.4%GPT-5Nearby rank
- 54.0%Kimi K2.7 Code
- 53.0%Grok 4.7Nearby rank
- 51.2%Gemini 3 Flash PreviewNearby rank
- 42.8%Grok 4.3Similar price
- 31.2%Qwen3.5 397B A17BSimilar price
Ahead of 2 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#55 of 161
- 94.8%Gemini 3.7 FlashSimilar price
- 91.7%DeepSeek V4 Pro 0813Similar price
- 88.3%
InklingNearby rank - 88.1%Mistral Large 4Nearby rank
- 87.9%Kimi K2.7 Code
- 87.9%Qwen3.7 PlusNearby rank
- 87.8%GLM 5Similar price
- 86.4%Qwen3.5 397B A17BSimilar price
Ahead of 2 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#38 of 158
- 98.6%DeepSeek V4 Pro 0813Similar price
- 97.2%Gemini 3.7 FlashSimilar price
- 95.6%Gemini 3 Flash PreviewNearby rank
- 95.6%Gemini 3.5 FlashNearby rank
- 95.6%Kimi K2.7 Code
- 95.6%Qwen3.7 MaxNearby rank
- 94.7%Claude Sonnet 5Nearby rank
- 88.9%Qwen3.5 397B A17BSimilar price
- 80.0%GLM 5Similar price
Ahead of 2 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#55 of 87
- 69.2%Gemini 3.7 FlashSimilar price
- 52.9%DeepSeek V4 Pro 0813Similar price
- 40.4%Qwen3 235B A22B Thinking 2507Similar price
- 40.3%InklingNearby rank
- 37.1%GPT-5.2Nearby rank
- 36.5%Kimi K2.7 Code
- 35.5%Claude Sonnet 4.6Nearby rank
- 34.9%Kimi K2.6Nearby rank
- 33.2%Grok 4.3Similar price
Ahead of 1 of 4 similarly priced models
DeepSWE 1.1
Primary coding-agent benchmark · 113 tasks · Data generated 2026-09-22
- pass@4
- 61.1%
- Average cost / attempt
- $2.82
- Benchmark runs
- 4
pass@1 is the success rate across scored attempts; pass@4 is the share of tasks solved in at least one of four attempts. Costs are reported by DeepSWE for each configuration. These results are not directly comparable to evaluations using different harnesses. Official source JSON.
All published scores
- Arena WebDevArena #771474
- default147495% CI 1465 to 1484 · 4,887 votes
- FrontierMath-Tier-4-v2-Private#57 of 7212.2%
- default12.2%
Source:Epoch AI evaluation - FrontierMath-Tiers-1-3-v2-Private#51 of 8754.0%
- default54.0%
Source:Epoch AI evaluation - GPQA diamond#55 of 16187.9%
- default87.9%
Source:Epoch AI evaluation - OTIS Mock AIME 2024-2025#38 of 15895.6%
- default95.6%
Source:Epoch AI evaluation - SimpleQA Verified#55 of 8736.5%
- default36.5%
Source:Epoch AI evaluation
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.