- Context
- 1Mtokens
- Listed
- 5d 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
Not ranked on Arena Text yet
Arena WebDev
Best Arena rating per model · Higher is better
Arena #1
- 1836
Claude Fable 5.2 - 1814Claude Opus 5.5Nearby rank
- 1773Claude Sonnet 5.5Nearby rank
- 1764
GPT-6 AstraSimilar price - 1742Claude Fable 5.1Predecessor
- 1649
Muse WonderSimilar price - 1615GPT-5.6 SolSimilar price
- 1512GPT-5.5Similar price
+94 vs Claude Fable 5.1 (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
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
Arena #1
- +0.304Claude Fable 5.2
- +0.195Claude Opus 5.5Nearby rank
- +0.170
Gemini 4 CarbonNearby rank - +0.148Claude Fable 5.1Predecessor
- +0.145GPT-6 AstraSimilar price
- +0.073Claude Fable 5Similar price
- +0.046GPT-5.6 SolSimilar price
- +0.018Muse WonderSimilar price
+0.155 vs Claude Fable 5.1 (predecessor) · Ahead of 4 of 4 similarly priced models
ARC-AGI-2
Best score per model · Higher is better
#4 of 85
- 95.0%GPT-6 AstraSimilar price
- 94.2%Gemini 4 CarbonNearby rank
- 94.2%GPT-6.1 SolNearby rank
- 93.8%Claude Fable 5.2
- 93.3%Claude Opus 5.5Nearby rank
- 92.5%GPT-5.6 SolSimilar price
- 90.0%Claude Fable 5.1Predecessor
- 89.2%Claude Fable 5Similar price
- 85.0%GPT-5.5Similar price
+3.8 pts vs Claude Fable 5.1 (predecessor) · Ahead of 3 of 4 similarly priced models
FrontierMath-Tier-4-v2-Private
Best score per model · Higher is better
#1 of 72
- 100.0%Claude Fable 5.2
- 100.0%GPT-6.1 SolNearby rank
- 97.6%GPT-6 AstraSimilar price
- 90.2%Claude Fable 5Similar price
- 87.8%Claude Fable 5.1Predecessor
- 82.9%GPT-5.6 SolSimilar price
- 72.5%GPT-5.5Similar price
+12.2 pts vs Claude Fable 5.1 (predecessor) · Ahead of 4 of 4 similarly priced models
FrontierMath-Tiers-1-3-v2-Private
Best score per model · Higher is better
#1 of 87
- 95.1%Claude Fable 5.2
- 93.7%GPT-6 AstraSimilar price
- 93.7%GPT-6.1 SolNearby rank
- 90.2%Claude Fable 5.1Predecessor
- 89.1%GPT-5.6 SolSimilar price
- 87.0%Claude Fable 5Similar price
- 85.3%GPT-5.5Similar price
+4.9 pts vs Claude Fable 5.1 (predecessor) · Ahead of 4 of 4 similarly priced models
GPQA diamond
Best score per model · Higher is better
#14 of 161
- 95.8%GPT-6 AstraSimilar price
- 94.1%Gemini 3.6 FlashNearby rank
- 94.0%
Grok 4.6Nearby rank - 93.9%Claude Fable 5.2
- 93.9%Claude Opus 5Similar price
- 93.5%GPT-5.6 SolSimilar price
- 90.7%GPT-5.5Similar price
- 85.9%Claude Fable 5Earlier version
+8.0 pts vs Claude Fable 5 (earlier version) · Ahead of 2 of 4 similarly priced models
OTIS Mock AIME 2024-2025
Best score per model · Higher is better
#13 of 158
- 100.0%Claude Fable 5Similar price
- 100.0%Claude Fable 5.1Predecessor
- 100.0%GPT-5.6 SolSimilar price
- 100.0%GPT-6 AstraSimilar price
- 100.0%GPT-6.1 TerraNearby rank
- 100.0%
Qwen3.8 Max (0902)Nearby rank - 99.7%Claude Fable 5.2
- 99.7%GPT-5.6 TerraNearby rank
- 99.4%
Qwen3.8 MaxNearby rank - 84.4%GPT-5.5Similar price
-0.3 pts vs Claude Fable 5.1 (predecessor) · Ahead of 1 of 4 similarly priced models
SimpleQA Verified
Best score per model · Higher is better
#1 of 87
- 94.0%Claude Fable 5.2
- 77.7%Gemini 4 CarbonNearby rank
- 76.0%Gemini 4 ArgonNearby rank
- 75.6%GPT-6 AstraSimilar price
- 70.8%Claude Fable 5.1Predecessor
- 70.7%Claude Fable 5Similar price
- 69.7%GPT-5.6 SolSimilar price
- 63.0%GPT-5.5Similar price
+23.2 pts vs Claude Fable 5.1 (predecessor) · 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 #11836
+94 vs Claude Fable 5.1 (predecessor)
- xhigh183695% CI 1813 to 1858 · 994 votes
- Agent ArenaArena #1+0.304
+0.155 vs Claude Fable 5.1 (predecessor)
- max+0.30495% CI +0.264 to +0.343 · 577,927 observations
- ARC-AGI-2#4 of 8593.8%
+3.8 pts vs Claude Fable 5.1 (predecessor)
- low86.7%
- medium89.6%
- high90.4%
- xhigh93.8%
- max92.5%
- FrontierMath-Tier-4-v2-Private#1 of 72100.0%
+12.2 pts vs Claude Fable 5.1 (predecessor)
- max100.0%
Source:Epoch AI evaluation - FrontierMath-Tiers-1-3-v2-Private#1 of 8795.1%
+4.9 pts vs Claude Fable 5.1 (predecessor)
- max95.1%
Source:Epoch AI evaluation - GPQA diamond#14 of 16193.9%
+8.0 pts vs Claude Fable 5 (earlier version)
- max93.9%
Source:Epoch AI evaluation - OTIS Mock AIME 2024-2025#13 of 15899.7%
-0.3 pts vs Claude Fable 5.1 (predecessor)
- max99.7%
Source:Epoch AI evaluation - SimpleQA Verified#1 of 8794.0%
+23.2 pts vs Claude Fable 5.1 (predecessor)
- max94.0%
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.