- input
- $1.75
- cached
- $0.175
- output
- $14.00
- Context
- 400Ktokens
- Released
- 11mo 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 #102
- 1487
Kimi K3Similar price - 1487
Gemini 3.1 Pro PreviewSimilar price - 1472
Claude Sonnet 5.5Similar price - 1455
GPT-5.1Predecessor - 1441GPT-6 LunaNearby rank
- 1437Gemma 4 26B A4BNearby rank
- 1437GPT-5.2
- 1436
MiniMax M3Nearby rank - 1436
DeepSeek V4 Flash 0423Nearby rank - 1354
Command ASimilar price
-18 vs GPT-5.1 (predecessor) · Ahead of 1 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Arena #96
- 1773Claude Sonnet 5.5Similar price
- 1676
Qwen3.8 Max (0902)Similar price - 1658Kimi K3Similar price
- 1446Gemini 3.1 Pro PreviewSimilar price
- 1435
GLM 4.7Nearby rank - 1434
MiMo-V2-ProNearby rank - 1420GPT-5.2
- 1417GPT-5Nearby rank
- 1414
InklingNearby rank - 1397GPT-5.1Predecessor
Level with GPT-5.1 (predecessor, overlapping 95% CIs) · Ahead of 0 of 4 similarly priced models
Arena Vision
Best Arena rating per model · Higher is better
Arena #62
- 1280
Grok 4.5Similar price - 1280Gemini 3.1 Pro PreviewSimilar price
- 1263Claude Sonnet 5.5Similar price
- 1251GPT-5.1Predecessor
- 1245
Step 5 PreviewNearby rank - 1243Qwen3.8 27BNearby rank
- 1242GPT-5.2
- 1242Gemma 4 26B A4BNearby rank
- 1242Grok 4.3Nearby rank
- 1199
Mistral Medium 3.5Similar price
Level with GPT-5.1 (predecessor, overlapping 95% CIs) · Ahead of 1 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 #42
- 1478Claude Sonnet 4.6Similar price
- 1469Claude Sonnet 5Similar price
- 1466Grok 4.5Similar price
- 1459Gemini 3.1 Pro PreviewSimilar price
- 1435MiniMax M3Nearby rank
- 1433Gemini 2.5 ProNearby rank
- 1425GPT-5.2
- 1419Claude Haiku 4.5Nearby rank
- 1416GPT-5.1Predecessor
Level with GPT-5.1 (predecessor, overlapping 95% CIs) · Ahead of 0 of 4 similarly priced models
Agent Arena
Best IPS score per model · Higher is better
Not ranked on Agent Arena yet
ARC-AGI-2
Best score per model · Higher is better
#41 of 85
- 83.9%Claude Sonnet 5.5Similar price
- 77.1%Gemini 3.1 Pro PreviewSimilar price
- 67.1%Grok 4.6Similar price
- 60.4%Kimi K3Similar price
- 59.3%GPT-6 LunaNearby rank
- 54.2%GPT-5.2 ProNearby rank
- 52.9%GPT-5.2
- 52.6%Grok 4.5Nearby rank
- 42.4%Qwen3.8 27BNearby rank
- 17.6%GPT-5.1Predecessor
+35.3 pts vs GPT-5.1 (predecessor) · Ahead of 0 of 4 similarly priced models
FrontierMath-2025-02-28-Private
Best score per model · Higher is better
#5 of 58
- 43.8%Claude Opus 4.7Nearby rank
- 40.7%Claude Opus 4.6Nearby rank
- 40.7%GPT-5.2
- 39.0%Gemini 3.5 FlashNearby rank
- 39.0%Kimi K2.6Nearby rank
- 36.9%Gemini 3.1 Pro PreviewSimilar price
- 32.4%Claude Sonnet 4.6Similar price
- 31.0%GPT-5.1Predecessor
- 15.2%Claude Sonnet 4.5Similar price
- 14.1%Gemini 2.5 ProSimilar price
+9.7 pts vs GPT-5.1 (predecessor) · Ahead of 4 of 4 similarly priced models
FrontierMath-Tier-4-2025-07-01-Private
Best score per model · Higher is better
#5 of 46
- 22.9%Claude Opus 4.7Nearby rank
- 22.9%Claude Opus 4.6Nearby rank
- 18.8%GPT-5.2
- 18.8%Gemini 3 Pro PreviewNearby rank
- 16.7%Gemini 3.1 Pro PreviewSimilar price
- 12.5%GPT-5.1Predecessor
- 8.3%Claude Sonnet 4.6Similar price
- 4.2%Claude Sonnet 4.5Similar price
- 4.2%Gemini 2.5 ProSimilar price
+6.3 pts vs GPT-5.1 (predecessor) · Ahead of 4 of 4 similarly priced models
FrontierMath-Tier-4-v2-Private
Best score per model · Higher is better
#34 of 72
- 80.5%Claude Sonnet 5.5Similar price
- 39.0%Kimi K3Similar price
- 34.1%Qwen3.8 Max (0902)Similar price
- 31.7%Claude Opus 4.7Nearby rank
- 31.7%Grok 4.6Nearby rank
- 31.7%GPT-5.2
- 29.3%Claude Sonnet 5Nearby rank
- 29.3%GLM 5.2Nearby rank
- 26.8%Gemini 3.1 Pro PreviewSimilar price
- 22.0%GPT-5Earlier version
+9.7 pts vs GPT-5 (earlier version) · Ahead of 1 of 4 similarly priced models
FrontierMath-Tiers-1-3-v2-Private
Best score per model · Higher is better
#32 of 87
- 88.8%Claude Sonnet 5.5Similar price
- 72.2%Kimi K3Similar price
- 68.8%GLM 5.3Nearby rank
- 68.4%Gemini 3.8 FlashNearby rank
- 67.4%GPT-5.2
- 67.4%DeepSeek V4.1 FlashNearby rank
- 66.0%Claude Opus 4.6Nearby rank
- 66.0%Grok 4.6Similar price
- 59.6%Gemini 3.1 Pro PreviewSimilar price
- 55.4%GPT-5Earlier version
+12.0 pts vs GPT-5 (earlier version) · Ahead of 2 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#30 of 161
- 95.6%Claude Sonnet 5.5Similar price
- 94.4%Gemini 3.1 Pro PreviewSimilar price
- 94.0%Grok 4.6Similar price
- 93.1%Kimi K3Similar price
- 91.7%DeepSeek V4 Pro 0813Nearby rank
- 91.6%GPT-5.6 LunaNearby rank
- 91.4%GPT-5.2
- 91.0%Claude Opus 4.8Nearby rank
- 91.0%DeepSeek V4 Flash 0731Nearby rank
- 87.6%GPT-5.1Predecessor
+3.8 pts vs GPT-5.1 (predecessor) · Ahead of 0 of 4 similarly priced models
Humanity's Last Exam
Best score per model · Higher is better
#13 of 34
- 46.4%Gemini 3.1 Pro PreviewSimilar price
- 46.1%Claude Sonnet 5.5Similar price
- 31.6%GPT-5 ProNearby rank
- 30.1%Claude Haiku 5.5Nearby rank
- 27.8%GPT-5.2
- 25.3%GPT-5Nearby rank
- 24.4%Kimi K2.5Nearby rank
- 23.7%GPT-5.1Predecessor
- 21.6%Gemini 2.5 Pro Preview 06-05Similar price
- 7.5%Claude Sonnet 4.5Similar price
+4.1 pts vs GPT-5.1 (predecessor) · Ahead of 2 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#35 of 158
- 100.0%Claude Sonnet 5.5Similar price
- 100.0%Qwen3.8 Max (0902)Similar price
- 97.2%Kimi K3Similar price
- 96.7%DeepSeek V4 Pro 0423Nearby rank
- 96.1%Kimi K2.6Nearby rank
- 96.1%GPT-5.2
- 95.6%Gemini 3.1 Pro PreviewSimilar price
- 95.6%Gemini 3 Flash PreviewNearby rank
- 88.6%GPT-5.1Predecessor
+7.5 pts vs GPT-5.1 (predecessor) · Ahead of 1 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#54 of 87
- 73.5%Gemini 3.1 Pro PreviewSimilar price
- 56.0%Grok 4.7Similar price
- 50.6%Kimi K3Similar price
- 48.0%GPT-5.1Predecessor
- 46.5%Claude Sonnet 5.5Similar price
- 40.4%Qwen3 235B A22B Thinking 2507Nearby rank
- 40.3%InklingNearby rank
- 37.1%GPT-5.2
- 36.5%Kimi K2.7 CodeNearby rank
- 35.5%Claude Sonnet 4.6Nearby rank
-10.9 pts vs GPT-5.1 (predecessor) · Ahead of 0 of 4 similarly priced models
SWE-bench Verified
Best score per model · Higher is better
#18 of 35
- 75.6%Gemini 3.1 Pro Preview Custom ToolsSimilar price
- 75.2%Claude Sonnet 4.6Similar price
- 74.2%GLM 5.1Nearby rank
- 73.8%Kimi K2.5Nearby rank
- 73.8%GPT-5.2
- 73.6%GPT-5Nearby rank
- 73.3%Claude Opus 4.1Nearby rank
- 71.3%Claude Sonnet 4.5Similar price
- 65.9%GPT-5.1Predecessor
- 57.6%Gemini 2.5 ProSimilar price
+7.9 pts vs GPT-5.1 (predecessor) · Ahead of 2 of 4 similarly priced models
Terminal Bench
Best score per model · Higher is better
#9 of 49
- 80.2%Gemini 3.1 Pro PreviewSimilar price
- 69.4%Gemini 3 Pro PreviewNearby rank
- 66.5%GPT-5.2-CodexNearby rank
- 64.9%GPT-5.2
- 64.3%Gemini 3 Flash PreviewNearby rank
- 63.1%Claude Opus 4.5Nearby rank
- 53.4%Claude Sonnet 4.6Similar price
- 47.6%GPT-5.1Predecessor
- 46.5%Claude Sonnet 4.5Similar price
- 32.6%Gemini 2.5 ProSimilar price
+17.3 pts vs GPT-5.1 (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.
Reasoning effort
ARC-AGI-2
Score by reasoning effort · Higher is better
FrontierMath-2025-02-28-Private
Score by reasoning effort · Higher is better
FrontierMath-Tier-4-2025-07-01-Private
Score by reasoning effort · Higher is better
GPQA diamond
Score by reasoning effort · Higher is better
OTIS Mock AIME 2024-2025
Score by reasoning effort · Higher is better
SimpleQA Verified
Score by reasoning effort · Higher is better
Terminal Bench
Score by reasoning effort · Higher is better
All published scores
- Arena TextArena #1021437
-18 vs GPT-5.1 (predecessor)
- high143795% CI 1434 to 1441 · 49,309 votes
- default143595% CI 1432 to 1439 · 82,951 votes
- Arena WebDevArena #961420
Level with GPT-5.1 (predecessor, overlapping 95% CIs)
- default142095% CI 1397 to 1443 · 1,065 votes
- Arena VisionArena #621242
Level with GPT-5.1 (predecessor, overlapping 95% CIs)
- high124295% CI 1236 to 1248 · 18,920 votes
- default122895% CI 1222 to 1234 · 20,001 votes
- Arena DocumentArena #421425
Level with GPT-5.1 (predecessor, overlapping 95% CIs)
- high142295% CI 1413 to 1431 · 7,552 votes
- default142595% CI 1419 to 1431 · 29,726 votes
- ARC-AGI-2#41 of 8552.9%
+35.3 pts vs GPT-5.1 (predecessor)
- low9.7%
- medium26.7%
- high43.3%
- xhigh52.9%
- unknown0.8%
- FrontierMath-2025-02-28-Private#5 of 5840.7%
+9.7 pts vs GPT-5.1 (predecessor)
- low26.6%
- medium36.9%
- high40.3%
- xhigh40.7%
Source:Epoch AI evaluation - FrontierMath-Tier-4-2025-07-01-Private#5 of 4618.8%
+6.3 pts vs GPT-5.1 (predecessor)
- low6.3%
- medium16.7%
- high18.8%
- xhigh18.8%
Source:Epoch AI evaluation - FrontierMath-Tier-4-v2-Private#34 of 7231.7%
+9.7 pts vs GPT-5 (earlier version)
- xhigh31.7%
Source:Epoch AI evaluation - FrontierMath-Tiers-1-3-v2-Private#32 of 8767.4%
+12.0 pts vs GPT-5 (earlier version)
- xhigh67.4%
Source:Epoch AI evaluation - GPQA diamond#30 of 16191.4%
+3.8 pts vs GPT-5.1 (predecessor)
- none73.2%
- low82.7%
- medium87.9%
- high88.2%
- xhigh91.4%
Source:Epoch AI evaluation - Humanity's Last Exam#13 of 3427.8%
+4.1 pts vs GPT-5.1 (predecessor)
- unknown27.8%
- OTIS Mock AIME 2024-2025#35 of 15896.1%
+7.5 pts vs GPT-5.1 (predecessor)
- none62.2%
- low78.9%
- medium93.9%
- high96.1%
- xhigh96.1%
Source:Epoch AI evaluation - SimpleQA Verified#54 of 8737.1%
-10.9 pts vs GPT-5.1 (predecessor)
- low32.8%
- medium32.7%
- high34.3%
- xhigh37.1%
Source:Epoch AI evaluation - SWE-bench Verified#18 of 3573.8%
+7.9 pts vs GPT-5.1 (predecessor)
- high73.8%
Source:Epoch AI evaluation - Terminal Bench#9 of 4964.9%
+17.3 pts vs GPT-5.1 (predecessor)
- medium64.9%DroidTerminal-Bench v2 LeaderboardOriginal-source terms apply
- medium62.9%Codex CLITerminal-Bench v2 LeaderboardOriginal-source terms apply
- unknown64.9%Droidhttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown62.9%Codex CLIhttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown60.7%Muxhttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown54.0%Terminus 2https://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
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.