- input
- $0.25
- cached
- $0.025
- output
- $2.00
- Context
- 400Ktokens
- Released
- 1y 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 #182
- 1416
Qwen3.5-122B-A10BSimilar price - 1405
Inkling SmallSimilar price - 1392
MiMo-V2-FlashNearby rank - 1391
MiniMax M2.5Nearby rank - 1390
GPT-5 Mini - 1388
Claude Sonnet 3.7Nearby rank - 1387
Solar Pro 4Nearby rank - 1383GPT-4.1 MiniPredecessor
- 1318
Llama 3.3 70B InstructSimilar price - 1290
Gemma 2 27BSimilar price
Level with GPT-4.1 Mini (predecessor, overlapping 95% CIs) · Ahead of 2 of 4 similarly priced models
Arena WebDev
Best Arena rating per model · Higher is better
Not ranked on Arena WebDev yet
Arena Vision
Best Arena rating per model · Higher is better
Arena #103
- 1234MiMo-V2.5Similar price
- 1226Qwen3.5-122B-A10BSimilar price
- 1208Inkling SmallSimilar price
- 1201GPT-4.1 MiniPredecessor
- 1200
Mistral Large 3 2512Similar price - 1189Gemini 2.5 Flash-Lite Preview 06-17Nearby rank
- 1188
Qwen3 VL 235B A22B ThinkingNearby rank - 1185GPT-5 Mini
- 1174Gemini 2.5 Flash-Lite Preview 09-2025Nearby rank
- 1172Gemini 2.0 FlashNearby rank
-16 vs GPT-4.1 Mini (predecessor) · Ahead of 0 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
Not ranked on Arena Document yet
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
#66 of 85
- 72.9%
DeepSeek V4.1 FlashSimilar price - 42.4%Qwen3.8 27BSimilar price
- 40.1%Inkling SmallSimilar price
- 10.3%Gemini 3.5 Flash LiteSimilar price
- 4.9%o3 ProNearby rank
- 4.9%
GLM 5Nearby rank - 4.4%GPT-5 Mini
- 4.0%Claude Haiku 4.5Nearby rank
- 4.0%DeepSeek V3.2Nearby rank
- 0.0%GPT-4.1 MiniPredecessor
+4.4 pts vs GPT-4.1 Mini (predecessor) · Ahead of 0 of 4 similarly priced models
FrontierMath-2025-02-28-Private
Best score per model · Higher is better
#17 of 58
- 31.0%GPT-5.1Nearby rank
- 28.3%GPT-5.4 MiniNearby rank
- 27.2%GPT-5 Mini
- 26.2%Qwen3.6 PlusSimilar price
- 25.9%GPT-5.4 NanoNearby rank
- 4.8%Gemini 2.5 FlashSimilar price
- 4.5%GPT-4.1 MiniPredecessor
- 3.8%GLM 4.6Similar price
- 0.3%Mistral Medium 3Similar price
+22.8 pts vs GPT-4.1 Mini (predecessor) · Ahead of 4 of 4 similarly priced models
FrontierMath-Tier-4-2025-07-01-Private
Best score per model · Higher is better
#16 of 46
- 8.3%Qwen3.6 PlusSimilar price
- 8.3%Claude Sonnet 4.6Nearby rank
- 6.3%GPT-5 Mini
- 6.3%GPT-5.4 NanoNearby rank
- 4.2%Qwen3.6 Max PreviewNearby rank
- 4.2%Gemini 2.5 FlashSimilar price
- 2.1%GLM 4.6Similar price
- 0.0%
Kimi K2 ThinkingSimilar price
Ahead of 3 of 4 similarly priced models
FrontierMath-Tier-4-v2-Private
Best score per model · Higher is better
#57 of 72
- 26.8%DeepSeek V4.1 FlashSimilar price
- 26.8%Mistral Large 4Similar price
- 17.1%Inkling SmallSimilar price
- 14.6%
Grok 4.3Nearby rank - 12.2%Kimi K2.7 CodeNearby rank
- 12.2%GPT-5 Mini
- 12.2%GPT-5.4 NanoNearby rank
- 9.8%GPT-5.4 MiniNearby rank
- 0.0%Gemini 3.5 Flash LiteSimilar price
Ahead of 1 of 4 similarly priced models
FrontierMath-Tiers-1-3-v2-Private
Best score per model · Higher is better
#55 of 87
- 67.4%DeepSeek V4.1 FlashSimilar price
- 51.2%Gemini 3 Flash PreviewNearby rank
- 51.2%GPT-5.4 MiniNearby rank
- 46.7%GPT-5 Mini
- 46.3%Inkling SmallSimilar price
- 45.3%DeepSeek V4 Pro 0423Nearby rank
- 38.2%Qwen3.6 PlusSimilar price
- 27.7%Gemini 3.1 Flash LiteSimilar price
- 6.7%GPT-4.1 MiniPredecessor
+40.0 pts vs GPT-4.1 Mini (predecessor) · Ahead of 3 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#100 of 161
- 88.5%Inkling SmallSimilar price
- 88.4%Qwen3.6 PlusSimilar price
- 75.8%Gemma 4 31BNearby rank
- 75.8%gpt-oss-120bNearby rank
- 75.0%GPT-5 Mini
- 73.2%Gemma 4 26B A4BNearby rank
- 72.6%Qwen3 MaxNearby rank
- 65.8%GPT-4.1 MiniPredecessor
- 47.4%Llama 3.3 70B InstructSimilar price
- 36.5%Gemma 2 27BSimilar price
+9.2 pts vs GPT-4.1 Mini (predecessor) · Ahead of 2 of 4 similarly priced models
Humanity's Last Exam
Best score per model · Higher is better
#19 of 34
- 24.4%Kimi K2.5Similar price
- 21.6%Gemini 2.5 Pro Preview 06-05Nearby rank
- 20.3%o3Nearby rank
- 19.4%GPT-5 Mini
- 14.2%Claude Opus 4.5Nearby rank
- 11.0%Gemini 2.5 Flash Preview 05-20Nearby rank
- 8.6%Gemini 3.1 Flash LiteSimilar price
- 5.7%Llama 4 MaverickSimilar price
- 4.5%Mistral Medium 3Similar price
Ahead of 3 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#66 of 158
- 93.3%Qwen3.6 PlusSimilar price
- 90.0%Inkling SmallSimilar price
- 86.7%Qwen3.5 PlusNearby rank
- 86.7%
Nemotron 3 UltraNearby rank - 86.7%GPT-5 Mini
- 86.7%Qwen3 235B A22B Thinking 2507Nearby rank
- 86.7%Qwen3.6 35B A3BNearby rank
- 44.7%GPT-4.1 MiniPredecessor
- 5.1%Llama 3.3 70B InstructSimilar price
- 1.4%Gemma 2 27BSimilar price
+41.9 pts vs GPT-4.1 Mini (predecessor) · Ahead of 2 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#73 of 87
- 44.1%Qwen3.6 PlusSimilar price
- 42.9%DeepSeek V4.1 FlashSimilar price
- 34.3%Kimi K2.5Similar price
- 25.4%Qwen3.5 PlusNearby rank
- 23.8%Claude Haiku 5.5Nearby rank
- 21.6%GPT-5 Mini
- 20.3%Qwen3.5-FlashNearby rank
- 20.0%Mistral Large 4Nearby rank
- 19.1%Inkling SmallSimilar price
- 12.7%GPT-4.1 MiniPredecessor
+8.9 pts vs GPT-4.1 Mini (predecessor) · Ahead of 1 of 4 similarly priced models
SWE-bench Verified
Best score per model · Higher is better
#29 of 35
- 77.6%DeepSeek V4 Pro 0423Similar price
- 75.4%Gemini 3 Flash PreviewSimilar price
- 73.8%Kimi K2.5Similar price
- 70.7%Claude Opus 4Nearby rank
- 65.9%GPT-5.1Nearby rank
- 64.7%GPT-5 Mini
- 62.3%o3Nearby rank
- 61.0%Claude Sonnet 3.7Nearby rank
- 57.9%Qwen3.6 PlusSimilar price
Ahead of 1 of 4 similarly priced models
Terminal Bench
Best score per model · Higher is better
#35 of 49
- 43.2%Kimi K2.5Similar price
- 42.7%MiniMax M2.5Similar price
- 35.7%Kimi K2 ThinkingNearby rank
- 35.5%Claude Haiku 4.5Nearby rank
- 34.8%GPT-5 Mini
- 33.4%GLM 4.7Nearby rank
- 32.6%Gemini 2.5 ProNearby rank
- 24.5%GLM 4.6Similar price
- 17.1%Gemini 2.5 FlashSimilar price
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
FrontierMath-Tiers-1-3-v2-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
Terminal Bench
Score by reasoning effort · Higher is better
All published scores
- Arena TextArena #1821390
Level with GPT-4.1 Mini (predecessor, overlapping 95% CIs)
- high139095% CI 1385 to 1395 · 26,718 votes
- Arena VisionArena #1031185
-16 vs GPT-4.1 Mini (predecessor)
- high118595% CI 1177 to 1193 · 28,904 votes
- ARC-AGI-2#66 of 854.4%
+4.4 pts vs GPT-4.1 Mini (predecessor)
- minimal1.7%
- low0.8%
- medium4.0%
- high4.4%
- FrontierMath-2025-02-28-Private#17 of 5827.2%
+22.8 pts vs GPT-4.1 Mini (predecessor)
- medium20.3%
- high27.2%
Source:Epoch AI evaluation - FrontierMath-Tier-4-2025-07-01-Private#16 of 466.3%
- medium4.2%
- high6.3%
Source:Epoch AI evaluation - FrontierMath-Tier-4-v2-Private#57 of 7212.2%
- high12.2%
Source:Epoch AI evaluation - FrontierMath-Tiers-1-3-v2-Private#55 of 8746.7%
+40.0 pts vs GPT-4.1 Mini (predecessor)
- minimal6.0%
- low18.2%
- high46.7%
Source:Epoch AI evaluation - GPQA diamond#100 of 16175.0%
+9.2 pts vs GPT-4.1 Mini (predecessor)
- minimal71.7%
- medium71.7%
- high75.0%
Source:Epoch AI evaluation - Humanity's Last Exam#19 of 3419.4%
- unknown19.4%
- OTIS Mock AIME 2024-2025#66 of 15886.7%
+41.9 pts vs GPT-4.1 Mini (predecessor)
- minimal55.6%
- medium78.3%
- high86.7%
Source:Epoch AI evaluation - SimpleQA Verified#73 of 8721.6%
+8.9 pts vs GPT-4.1 Mini (predecessor)
- high21.6%
Source:Epoch AI evaluation - SWE-bench Verified#29 of 3564.7%
- medium64.7%
Source:Epoch AI evaluation - Terminal Bench#35 of 4934.8%
- medium31.9%Codex CLITerminal-Bench v2 LeaderboardOriginal-source terms apply
- medium29.2%OpenHandsTerminal-Bench v2 LeaderboardOriginal-source terms apply
- medium24.0%Terminus 2Terminal-Bench v2 LeaderboardOriginal-source terms apply
- medium22.2%Mini-SWE-AgentTerminal-Bench v2 LeaderboardOriginal-source terms apply
- unknown34.8%spoox-o-mhttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown34.8%spoox-mExternal evaluation via Epoch AIOriginal-source terms apply
- unknown31.9%Codex CLIhttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown29.2%OpenHandshttps://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown24.0%Terminus 2https://www.tbench.ai/leaderboard/terminal-bench/2.0Original-source terms apply
- unknown22.2%Mini-SWE-Agenthttps://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.