Claude Fable 5.1 vs GPT-6.2 Astra
GPT-6.2 Astra leads by 3.0 capability points (≈95% interval +0.3 to +5.7): a clear lead over Claude Fable 5.1 · P(higher) ≈ 98%
Wins 43 of 68 shared benchmarks (24 lost, 1 tied)
Capability vs. relative cost
Claude Fable 5.1GPT-6.2 AstraClosed weightsOpen weights
Benchmark wins24 – 43 · 68 shared, 1 tied
Comparing matching setups only (35 benchmarks excluded) · suites counted once
Area← Claude Fable 5.1GPT-6.2 Astra →
Data: Artificial Analysis, Surge AI, Epoch AI and 16 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.2 Astra minus Claude Fable 5.1.
| AA-PersonalityBench Informativeness index | Language & writing › Style & personality | 96.2 | 31.3 | −65.5 |
| AA-PersonalityBench Transparency index | Language & writing › Style & personality | 98.1 | 40.9 | −57.8 |
| SpireClimb | Games | 2.0 | 41.3 | +39.3 |
| AA-PersonalityBench Organization index | Language & writing › Style & personality | 60.0 | 21.8 | −38.6 |
| Mystery Game Puzzles | Games | 58.0 | 95.0 | +37.0 |
| EBR-bench v4 (card ban, single agent) | Games | 57.1 | 92.4 | +35.2 |
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 77.4 | 43.6 | −33.8 |
| AA-PersonalityBench Expressiveness index | Language & writing › Style & personality | 98.1 | 66.5 | −31.9 |
| Chess Puzzles (task version 2.0) | Games | 48.0 | 78.0 | +30.0 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 22.4 | 52.4 | +30.0 |
| AA-PersonalityBench Conviction index | Language & writing › Style & personality | 100 | 71.3 | −29.0 |
| AA-PersonalityBench Warmth index | Language & writing › Style & personality | 94.3 | 68.1 | −26.5 |
| MirrorCode (ML, +Private, 2L) | Software › Software engineering | 93.3 | 67.8 | −25.6 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 52.6 | 77.1 | +24.5 |
| AA-3D v1.0 Elo | Software › Frontend & design | 1139 | 1318 | +24.3 |
| AA-PersonalityBench Imagination index | Language & writing › Style & personality | 86.7 | 63.3 | −23.6 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 18.3 | 39.4 | +21.1 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 77.9 | 57.9 | −20.0 |
| Blueprint-Bench 2 | Perception & media › Spatial & visual reasoning | 0.4 | 0.6 | +19.5 |
| Furniture Assembly Benchmark | Perception & media › Spatial & visual reasoning | 70.0 | 89.2 | +19.2 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 22.5 | 41.4 | +18.9 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 30.3 | 47.1 | +16.8 |
| AA-PersonalityBench Slop index | Language & writing › Style & personality | 23.8 | 7.4 | −16.6 |
| SWE Atlas v1.1 - Overall | Software › Software engineering | 76.5 | 60.5 | −16.0 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 51.3 | 66.9 | +15.6 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 27.2 | 42.0 | +14.8 |
| FrontierMath Tier 4 (v2) - Private | Math | 87.8 | 100.0 | +12.2 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 63.8 | 75.0 | +11.2 |
| Chartography (November 2026 rescoring) | Perception & media › Documents & charts | 79.6 | 90.7 | +11.1 |
| FrontierMath Erdős | Math | 0.0 | 10.3 | +10.3 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 36.5 | 45.8 | +9.3 |
| SimpleBench | Knowledge & reasoning › General knowledge | 86.6 | 77.3 | −9.3 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 60.4 | 69.6 | +9.2 |
| Harvey LAB-AA v1.1 (hallucination-gated) | Professional › Legal | 6.4 | 15.0 | +8.6 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 51.6 | 43.1 | −8.5 |
| HealthBench Professional (Length-Adjusted) | Health & medicine | 58.1 | 66.3 | +8.2 |
| SimpleQA Verified (Epoch anti-abstention) | Knowledge & reasoning › General knowledge | 70.8 | 78.5 | +7.7 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 85.1 | 78.2 | −6.9 |
| GDP.xlsx | Professional › General knowledge work | 23.1 | 29.7 | +6.6 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 32.2 | 38.3 | +6.1 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 77.7 | 83.7 | +6.0 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 90.0 | 95.8 | +5.8 |
| MysteryMechanism | Math | 47.7 | 53.2 | +5.4 |
| AA-AnalystAgent (pass^5) | Professional › General knowledge work | 57.5 | 62.5 | +5.0 |
| RiemannBench (November 2026 rescoring) | Math | 71.2 | 76.0 | +4.8 |
| Epoch Capabilities Index (ECI) | Composite indexes | 167 | 173 | +4.6 |
| LLM Debate Benchmark | Language & writing › Language & wordplay | 1706 | 1673 | −4.6 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 28.1 | 37.0 | +4.4 |
| MMMU Pro | Knowledge & reasoning › General knowledge | 86.1 | 89.5 | +3.4 |
| GPQA Diamond | Science › Multidisciplinary | 93.7 | 97.0 | +3.3 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 31.2 | 34.4 | +3.2 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 13.7 | 15.2 | +3.0 |
| Dual-Arm Sim (task completion) | Tool & computer use › Robotics | 0.0 | 2.5 | +2.5 |
| FrontierMath Tiers 1-3 (v2) - Private | Math | 90.2 | 92.6 | +2.5 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 42.0 | 44.0 | +2.0 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1671 | 1644 | −1.5 |
| Gray Swan IPI Attack Success (k=15) | Safety & alignment | 1.0 | 2.4 | −1.4 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 47.1 | 45.8 | −1.3 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 97.5 | 98.5 | +1.0 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 84.0 | 84.9 | +0.9 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 14.2 | 12.4 | −0.9 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | 9.3 | 7.7 | −0.8 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 76.7 | 76.0 | −0.7 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 9.6 | 9.2 | −0.4 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 13.5 | 14.0 | +0.2 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 49.4 | 49.2 | −0.2 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.0 | 1.2 | +0.1 |
| OTIS Mock AIME 2024-2025 | Math | 100.0 | 100.0 | 0.0 |
Data: Artificial Analysis, Surge AI, Epoch AI and 16 othersatlas.village.dev