- input
- $0.08
- cached
- $0.04
- output
- $0.35
- Context
- 131Ktokens
- Released
- 1y ago
- Size
- 27Bparameters
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 #205
- 1393
Step 3.5 FlashSimilar price - 1382
Qwen3 30B A3B Instruct 2507Similar price - 1373
GLM 4.5 AirNearby rank - 1369
Qwen3-Next 80B-A3B (Thinking)Nearby rank - 1365
Gemma 3 27B - 1364GLM 4.7 FlashSimilar price
- 1360Gemini 2.0 FlashNearby rank
- 1322
Llama 4 ScoutSimilar price - 1290Gemma 2 27BPredecessor
+76 vs Gemma 2 27B (predecessor) · Ahead of 1 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 #113
- 1248
MiMo-V2.6-FlashSimilar price - 1161
Claude Sonnet 3.5 v2Nearby rank - 1161GLM 4.6VNearby rank
- 1160Gemma 3 27B
- 1160
Mistral Medium 3.1Nearby rank - 1155Mistral Medium 3Nearby rank
- 1147
GPT-5 NanoSimilar price - 1128Llama 4 ScoutSimilar price
- 1019
Nova Lite 1.0Similar price
Ahead of 2 of 4 similarly priced models
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
GPQA diamond
Best score per model · Higher is better
#141 of 161
- 69.4%GPT-5 NanoSimilar price
- 63.8%Qwen3 14BSimilar price
- 60.5%GLM 4.7 FlashSimilar price
- 51.8%Llama 4 ScoutSimilar price
- 48.1%Qwen-PlusNearby rank
- 47.9%GPT-4o (2024-11-20)Nearby rank
- 47.7%Gemma 3 27B
- 47.6%Magistral Small 1.2Nearby rank
- 47.4%Llama 3.3 70B InstructNearby rank
- 36.5%Gemma 2 27BPredecessor
+11.2 pts vs Gemma 2 27B (predecessor) · Ahead of 0 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#131 of 158
- 81.1%GPT-5 NanoSimilar price
- 66.4%Qwen3 14BSimilar price
- 58.3%GLM 4.7 FlashSimilar price
- 28.9%GPT-4.1 NanoNearby rank
- 28.1%Magistral Small 1.2Nearby rank
- 22.5%Gemma 3 27B
- 17.8%Qwen-PlusNearby rank
- 16.7%Gemma 3 12BNearby rank
- 7.8%Llama 4 ScoutSimilar price
- 1.4%Gemma 2 27BPredecessor
+21.1 pts vs Gemma 2 27B (predecessor) · Ahead of 1 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.
All published scores
- Arena TextArena #2051365
+76 vs Gemma 2 27B (predecessor)
- default136595% CI 1362 to 1369 · 46,613 votes
- Arena VisionArena #1131160
- default116095% CI 1152 to 1168 · 16,943 votes
- GPQA diamond#141 of 16147.7%
+11.2 pts vs Gemma 2 27B (predecessor)
- default47.7%
Source:Epoch AI evaluation - OTIS Mock AIME 2024-2025#131 of 15822.5%
+21.1 pts vs Gemma 2 27B (predecessor)
- default22.5%
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.