- input
- $0.15
- cached
- $0.003
- output
- $0.60
- Context
- 1Mtokens
- Listed
- 3w ago
- Size
- 763Bparameters
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
Not ranked on Arena Text yet
Arena WebDev
Best Arena rating per model · Higher is better
Arena #17
- 1673
GPT-6 SolNearby rank - 1670
Qwen3.8 MaxNearby rank - 1664
DeepSeek V4.1 Flash 1017 - 1658
Kimi K3Nearby rank - 1655
Muse Spark 1.3Nearby rank - 1609Muse EmberSimilar price
- 1370
Gemma 4 31BSimilar price - 1369
Solar Pro 4Similar price - 1362DeepSeek V3.2Predecessor
- 1358
Hy3 previewSimilar price
+301 vs DeepSeek V3.2 (predecessor) · Ahead of 4 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
Arena #22
- +0.023Gemini 3.8 FlashNearby rank
- +0.022Muse Spark 1.3Nearby rank
- +0.020DeepSeek V4.1 Flash 1017
- +0.020
Claude Sonnet 5Nearby rank - +0.018Muse WonderNearby rank
- +0.017Muse EmberSimilar price
- +0.016GPT-6 LunaSimilar price
- +0.005Claude Haiku 5.5Similar price
- -0.226Solar Pro 4Similar price
Ahead of 1 of 4 similarly priced models
ARC-AGI-2
Best score per model · Higher is better
#20 of 85
- 83.3%GPT-5.4 ProNearby rank
- 81.7%
Inkling 2 SmallNearby rank - 78.1%DeepSeek V4.1 Flash 1017
- 77.1%Gemini 3.1 Pro PreviewNearby rank
- 75.8%Claude Opus 4.7Nearby rank
- 61.3%Claude Haiku 5.5Similar price
- 4.0%DeepSeek V3.2Predecessor
- 1.3%
Qwen3 235B A22B Instruct 2507Similar price - 0.0%Llama 4 ScoutSimilar price
- 0.0%GPT-4o-mini (2024-07-18)Similar price
+74.0 pts vs DeepSeek V3.2 (predecessor) · Ahead of 4 of 4 similarly priced models
FrontierMath-Tier-4-v2-Private
Best score per model · Higher is better
#23 of 72
- 61.0%GPT-5.6 LunaSimilar price
- 56.1%GPT-6 LunaSimilar price
- 46.3%Qwen3.8 MaxNearby rank
- 46.3%Claude Haiku 5.5Similar price
- 46.3%DeepSeek V4.1 Flash 1017
- 46.3%Muse Spark 1.3Nearby rank
- 46.0%GPT-5.2 ProNearby rank
- 29.3%
GLM 5.3Similar price
Ahead of 1 of 4 similarly priced models
FrontierMath-Tiers-1-3-v2-Private
Best score per model · Higher is better
#28 of 87
- 75.1%Claude Haiku 5.5Similar price
- 72.2%Kimi K3Nearby rank
- 71.6%Gemini 3.7 FlashNearby rank
- 71.2%DeepSeek V4.1 Flash 1017
- 70.2%Claude Opus 4.7Nearby rank
- 68.8%GLM 5.3Similar price
- 20.4%Qwen3.6 35B A3BSimilar price
- 0.7%GPT-4o-mini (2024-07-18)Similar price
Ahead of 3 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#20 of 161
- 93.3%GPT-5.6 TerraNearby rank
- 93.3%GPT-5.4Nearby rank
- 93.2%DeepSeek V4.1 Flash 1017
- 93.1%Kimi K3Nearby rank
- 92.8%Gemini 3.5 FlashNearby rank
- 89.6%Claude Haiku 5.5Similar price
- 84.8%Qwen3.6 35B A3BSimilar price
- 75.8%Gemma 4 31BSimilar price
- 75.8%gpt-oss-120bSimilar price
Ahead of 4 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#15 of 158
- 99.7%GPT-5.6 TerraNearby rank
- 99.4%Qwen3.8 MaxNearby rank
- 99.4%DeepSeek V4.1 Flash 1017
- 99.2%Muse Spark 1.3Nearby rank
- 99.2%
Grok 4.6Nearby rank - 98.9%Claude Haiku 5.5Similar price
- 88.9%gpt-oss-120bSimilar price
- 86.7%Qwen3.6 35B A3BSimilar price
- 73.3%Gemma 4 31BSimilar price
Ahead of 4 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#48 of 87
- 43.2%GPT-5.6 TerraNearby rank
- 42.9%DeepSeek V4.1 FlashNearby rank
- 42.5%DeepSeek V4.1 Flash 1017
- 41.4%GPT-6 LunaNearby rank
- 41.0%GPT-5.6 LunaNearby rank
- 41.0%GLM 5.3Similar price
- 23.8%Claude Haiku 5.5Similar price
- 10.4%Gemma 4 31BSimilar price
- 8.3%GPT-4o-mini (2024-07-18)Similar price
Ahead of 4 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.
Reasoning effort
ARC-AGI-2
Score by reasoning effort · Higher is better
All published scores
- Arena WebDevArena #171664
+301 vs DeepSeek V3.2 (predecessor)
- max166495% CI 1657 to 1670 · 13,952 votes
- Agent ArenaArena #22+0.020
- max+0.02095% CI +0.015 to +0.025 · 24,468,135 observations
- ARC-AGI-2#20 of 8578.1%
+74.0 pts vs DeepSeek V3.2 (predecessor)
- none2.1%
- low73.6%
- high68.6%
- max78.1%
- FrontierMath-Tier-4-v2-Private#23 of 7246.3%
- max46.3%
Source:Epoch AI evaluation - FrontierMath-Tiers-1-3-v2-Private#28 of 8771.2%
- max71.2%
Source:Epoch AI evaluation - GPQA diamond#20 of 16193.2%
- max93.2%
Source:Epoch AI evaluation - OTIS Mock AIME 2024-2025#15 of 15899.4%
- max99.4%
Source:Epoch AI evaluation - SimpleQA Verified#48 of 8742.5%
- max42.5%
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.