- input
- $1.25
- cached
- $0.125
- output
- $10.00
- Context
- 400Ktokens
- Released
- 12mo 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 #71
- 1472
Claude Sonnet 5.5Similar price - 1466
Grok 4.5Similar price - 1456Grok 4.6Nearby rank
- 1456Claude Sonnet 4.5Nearby rank
- 1455
GPT-5.1 - 1455
Qwen3.7 PlusNearby rank - 1455
Gemini 3.5 Flash LiteNearby rank - 1446Gemini 2.5 ProSimilar price
- 1435GPT-5Predecessor
- 1427
Mistral Medium 3.5Similar price
+21 vs GPT-5 (predecessor) · Ahead of 2 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Arena #108
- 1676Qwen3.8 Max (0902)Similar price
- 1641Grok 4.7Similar price
- 1417GPT-5Predecessor
- 1398
MiniMax M2.7Nearby rank - 1398GPT-5.4 MiniNearby rank
- 1397GPT-5.1
- 1393Claude Sonnet 4.5Nearby rank
- 1389Claude Opus 4.1Nearby rank
- 1263Mistral Medium 3.5Similar price
- 1228Gemini 2.5 ProSimilar price
Level with GPT-5 (predecessor, overlapping 95% CIs) · Ahead of 2 of 4 similarly priced models
Arena Vision
Best Arena rating per model · Higher is better
Arena #52
- 1280Grok 4.5Similar price
- 1263Claude Sonnet 5.5Similar price
- 1255GPT-5.6 LunaNearby rank
- 1252
Kimi K2.5Nearby rank - 1251GPT-5.1
- 1250GPT-5.4 MiniNearby rank
- 1249Grok 4.7Nearby rank
- 1247Gemini 2.5 ProSimilar price
- 1211GPT-5Predecessor
- 1199Mistral Medium 3.5Similar price
+40 vs GPT-5 (predecessor) · 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 #45
- 1480
Muse Spark 1.1Similar price - 1469Claude Sonnet 5Similar price
- 1466Grok 4.5Similar price
- 1433Gemini 2.5 ProSimilar price
- 1425GPT-5.2Nearby rank
- 1419Claude Haiku 4.5Nearby rank
- 1416GPT-5.1
- 1406
GLM 5V TurboNearby rank
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
#52 of 85
- 83.9%Claude Sonnet 5.5Similar price
- 67.1%Grok 4.6Similar price
- 60.4%Kimi K3Similar price
- 18.9%GPT-5.4 MiniNearby rank
- 18.3%GPT-5 ProNearby rank
- 17.6%GPT-5.1
- 13.6%Claude Sonnet 4.5Nearby rank
- 11.8%Kimi K2.5Nearby rank
- 9.9%GPT-5Predecessor
- 4.9%Gemini 2.5 ProSimilar price
+7.8 pts vs GPT-5 (predecessor) · Ahead of 1 of 4 similarly priced models
FrontierMath-2025-02-28-Private
Best score per model · Higher is better
#15 of 58
- 33.4%GLM 5.1Similar price
- 32.4%GPT-5Predecessor
- 32.4%Claude Sonnet 4.6Nearby rank
- 31.0%GPT-5.1
- 28.3%GPT-5.4 MiniNearby rank
- 27.2%GPT-5 MiniNearby rank
- 14.1%Gemini 2.5 ProSimilar price
- 10.3%Gemini 2.5 Pro Preview 06-05Similar price
- 5.9%Claude Haiku 4.5Similar price
-1.4 pts vs GPT-5 (predecessor) · Ahead of 3 of 4 similarly priced models
FrontierMath-Tier-4-2025-07-01-Private
Best score per model · Higher is better
#11 of 46
- 14.6%Kimi K2.6Nearby rank
- 12.5%GPT-5Predecessor
- 12.5%GPT-5.1
- 12.5%GLM 5.1Similar price
- 8.3%Qwen3.6 PlusNearby rank
- 4.2%Gemini 2.5 ProSimilar price
- 2.1%Claude Haiku 4.5Similar price
- 2.1%Gemini 2.5 Pro Preview 06-05Similar price
Level with GPT-5 (predecessor) · Ahead of 3 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#58 of 161
- 94.0%Grok 4.6Similar price
- 92.3%Qwen3.8 Max (0902)Similar price
- 87.9%Qwen3.7 PlusNearby rank
- 87.8%GLM 5Nearby rank
- 87.6%GPT-5.1
- 87.4%
Qwen3.6 Max PreviewNearby rank - 87.4%Claude Sonnet 4.6Nearby rank
- 86.2%GPT-5Predecessor
- 85.3%Gemini 2.5 ProSimilar price
- 49.0%Mistral Large 2407Similar price
+1.5 pts vs GPT-5 (predecessor) · Ahead of 2 of 4 similarly priced models
Humanity's Last Exam
Best score per model · Higher is better
#16 of 34
- 46.4%Gemini 3.1 Pro PreviewSimilar price
- 46.1%Claude Sonnet 5.5Similar price
- 25.3%GPT-5Predecessor
- 24.4%Kimi K2.5Nearby rank
- 23.7%GPT-5.1
- 21.6%Gemini 2.5 Pro Preview 06-05Similar price
- 20.3%o3Nearby rank
- 7.5%Claude Sonnet 4.5Similar price
-1.6 pts vs GPT-5 (predecessor) · Ahead of 2 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#63 of 158
- 100.0%Qwen3.8 Max (0902)Similar price
- 99.2%Grok 4.6Similar price
- 91.4%GPT-5Predecessor
- 88.9%Qwen3.5 397B A17BNearby rank
- 88.9%
InklingNearby rank - 88.6%GPT-5.1
- 87.8%
DeepSeek ReasonerNearby rank - 87.8%GPT-5.4 NanoNearby rank
- 84.7%Gemini 2.5 ProSimilar price
- 8.5%Mistral Large 2407Similar price
-2.8 pts vs GPT-5 (predecessor) · Ahead of 2 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#35 of 87
- 66.2%Gemini 3.5 FlashSimilar price
- 56.0%Grok 4.7Similar price
- 50.1%GPT-5Predecessor
- 48.7%Qwen3 MaxNearby rank
- 48.3%Grok 4.5Nearby rank
- 48.0%GPT-5.1
- 47.3%Qwen3.8 Max (0902)Similar price
- 47.0%Claude Opus 4.6Nearby rank
- 46.5%Claude Sonnet 5.5Similar price
-2.1 pts vs GPT-5 (predecessor) · Ahead of 2 of 4 similarly priced models
SWE-bench Verified
Best score per model · Higher is better
#28 of 35
- 77.3%Qwen3.7 MaxSimilar price
- 75.2%Claude Sonnet 4.6Similar price
- 74.2%GLM 5.1Similar price
- 73.6%GPT-5Predecessor
- 71.3%Claude Sonnet 4.5Nearby rank
- 70.7%Claude Opus 4Nearby rank
- 65.9%GPT-5.1
- 64.7%GPT-5 MiniNearby rank
- 62.3%o3Nearby rank
- 57.6%Gemini 2.5 ProSimilar price
-7.6 pts vs GPT-5 (predecessor) · Ahead of 1 of 4 similarly priced models
Terminal Bench
Best score per model · Higher is better
#20 of 49
- 80.2%Gemini 3.1 Pro PreviewSimilar price
- 53.4%Claude Sonnet 4.6Similar price
- 52.4%GLM 5Nearby rank
- 49.6%GPT-5Predecessor
- 47.6%GPT-5.1
- 46.5%Claude Sonnet 4.5Nearby rank
- 45.1%MiniMax M2.7Nearby rank
- 35.5%Claude Haiku 4.5Similar price
- 32.6%Gemini 2.5 ProSimilar price
-2.0 pts vs GPT-5 (predecessor) · Ahead of 2 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
Humanity's Last Exam
Score by reasoning effort · Higher is better
OTIS Mock AIME 2024-2025
Score by reasoning effort · Higher is better
Terminal Bench
Score by reasoning effort · Higher is better
All published scores
- Arena TextArena #711455
+21 vs GPT-5 (predecessor)
- high145595% CI 1452 to 1459 · 41,405 votes
- default143995% CI 1435 to 1442 · 44,207 votes
- Arena WebDevArena #1081397
Level with GPT-5 (predecessor, overlapping 95% CIs)
- medium139795% CI 1386 to 1409 · 4,857 votes
- default134295% CI 1334 to 1351 · 10,598 votes
- Arena VisionArena #521251
+40 vs GPT-5 (predecessor)
- high125195% CI 1243 to 1259 · 9,794 votes
- default124095% CI 1233 to 1248 · 10,753 votes
- Arena DocumentArena #451416
- default141695% CI 1407 to 1425 · 8,736 votes
- ARC-AGI-2#52 of 8517.6%
+7.8 pts vs GPT-5 (predecessor)
- none0.4%
- low1.9%
- medium6.5%
- high17.6%
- FrontierMath-2025-02-28-Private#15 of 5831.0%
-1.4 pts vs GPT-5 (predecessor)
- none2.1%
- low17.3%
- medium26.9%
- high31.0%
Source:Epoch AI evaluation - FrontierMath-Tier-4-2025-07-01-Private#11 of 4612.5%
Level with GPT-5 (predecessor)
- medium4.2%
- high12.5%
Source:Epoch AI evaluation - GPQA diamond#58 of 16187.6%
+1.5 pts vs GPT-5 (predecessor)
- none66.7%
- medium85.0%
- high87.6%
Source:Epoch AI evaluation - Humanity's Last Exam#16 of 3423.7%
-1.6 pts vs GPT-5 (predecessor)
- none6.8%
- unknown23.7%
- OTIS Mock AIME 2024-2025#63 of 15888.6%
-2.8 pts vs GPT-5 (predecessor)
- none37.8%
- low63.9%
- medium85.6%
- high88.6%
Source:Epoch AI evaluation - SimpleQA Verified#35 of 8748.0%
-2.1 pts vs GPT-5 (predecessor)
- high48.0%
Source:Epoch AI evaluation - SWE-bench Verified#28 of 3565.9%
-7.6 pts vs GPT-5 (predecessor)
- high65.9%
Source:Epoch AI evaluation - Terminal Bench#20 of 4947.6%
-2.0 pts vs GPT-5 (predecessor)
- medium47.6%Terminus 2Terminal-Bench v2 LeaderboardOriginal-source terms apply
- unknown47.6%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.