Claude Haiku 5.5 vs GPT-6.2 Sol
GPT-6.2 Sol leads by 5.7 capability points (≈95% interval +3.3 to +8.2): a clear lead over Claude Haiku 5.5 · P(higher) ≈ 100%
Wins 51 of 63 shared benchmarks (11 lost, 1 tied)
Capability vs. relative cost
Claude Haiku 5.5GPT-6.2 SolClosed weightsOpen weights
Benchmark wins11 – 51 · 63 shared, 1 tied
Comparing matching setups only (10 benchmarks excluded) · suites counted once
Area← Claude Haiku 5.5GPT-6.2 Sol →
Data: Artificial Analysis, Vals AI, Arena and 9 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.2 Sol minus Claude Haiku 5.5.
| ARC-AGI-3 (Semi-Private) | Games | 2.8 | 98.2 | +95.4 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 10.0 | 58.6 | +48.6 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 20.0 | 67.6 | +47.6 |
| AA-PersonalityBench Informativeness index | Language & writing › Style & personality | 75.3 | 37.7 | −37.9 |
| AA-3D v1.0 Elo | Software › Frontend & design | 998 | 1259 | +35.5 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 36.0 | 71.3 | +35.3 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 50.2 | 85.3 | +35.1 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 29.6 | 64.0 | +34.4 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 12.0 | 41.3 | +29.2 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 4.0 | 32.0 | +28.0 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 32.8 | 60.1 | +27.3 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 61.3 | 85.8 | +24.6 |
| Terminal-Bench 4.0 | Software › Software engineering | 36.4 | 60.6 | +24.2 |
| Blueprint-Bench 2 | Perception & media › Spatial & visual reasoning | 0.3 | 0.6 | +24.0 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | -0.6 | 11.3 | +23.2 |
| AA-PersonalityBench Expressiveness index | Language & writing › Style & personality | 77.2 | 55.3 | −22.1 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 36.5 | 57.3 | +20.8 |
| AA-PersonalityBench Transparency index | Language & writing › Style & personality | 77.2 | 56.9 | −20.5 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 39.1 | 59.4 | +20.3 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 52.8 | 72.4 | +19.6 |
| Vals Science Index | Science › Multidisciplinary | 38.5 | 57.8 | +19.3 |
| Artificial Analysis Engineering Index | Science › Engineering | 42.5 | 61.2 | +18.6 |
| Artificial Analysis Legal Index | Professional › Legal | 38.7 | 55.0 | +16.4 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 37.5 | 53.3 | +15.8 |
| Artificial Analysis Economics Index | Composite indexes | 42.7 | 57.4 | +14.7 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | -5.5 | 23.3 | +14.4 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 12.3 | 40.4 | +14.0 |
| VIALS-AA (pass^3) | Science › Life sciences | 49.1 | 62.7 | +13.7 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 43.6 | 57.2 | +13.6 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 66.3 | 78.3 | +12.0 |
| AA-PersonalityBench Conviction index | Language & writing › Style & personality | 67.6 | 79.2 | +11.7 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 68.9 | 80.3 | +11.4 |
| SimpleBench | Knowledge & reasoning › General knowledge | 59.2 | 69.9 | +10.7 |
| AA-PersonalityBench Warmth index | Language & writing › Style & personality | 69.5 | 60.1 | −9.6 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 34.9 | 44.4 | +9.4 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 20.8 | 30.0 | +9.2 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 65.5 | 74.3 | +8.8 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 89.0 | 97.5 | +8.5 |
| VIALS-AA (average of 3) | Science › Life sciences | 60.5 | 68.9 | +8.5 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 76.2 | 84.2 | +8.0 |
| HLAB-Vals-1.1 | Professional › Legal | 4.2 | 12.1 | +7.9 |
| BioMysteryBench (Vals Terminus 2) | Science › Life sciences | 70.4 | 77.8 | +7.4 |
| Vals Index v2.2 | Composite indexes | 58.7 | 66.1 | +7.3 |
| Legal Research Bench | Professional › Legal | 43.8 | 37.0 | −6.7 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 65.4 | 58.9 | −6.5 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 0.8 | 13.4 | +6.3 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 0.2 | 10.9 | +5.3 |
| AA-PersonalityBench Slop index | Language & writing › Style & personality | 10.5 | 15.4 | +4.9 |
| AA-PersonalityBench Organization index | Language & writing › Style & personality | 41.0 | 45.7 | +4.8 |
| MultiChallenge | Language & writing › Conversation | 67.7 | 72.0 | +4.3 |
| AA-PersonalityBench Imagination index | Language & writing › Style & personality | 63.8 | 60.1 | −3.8 |
| Vals Legal Index | Professional › Legal | 70.3 | 66.7 | −3.6 |
| VIALS-AA (pass@3) | Science › Life sciences | 71.4 | 74.5 | +3.1 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 90.4 | 87.6 | −2.8 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 35.0 | 37.2 | +2.2 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1572 | 1612 | +2.2 |
| MysteryMechanism | Math | 35.1 | 36.9 | +1.8 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | -1.5 | 1.2 | +1.4 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | 6.6 | 4.3 | −1.1 |
| SWE Atlas v1.1 - Overall | Software › Software engineering | 59.4 | 60.5 | +1.0 |
| Vals Finance Index | Professional › Finance, accounting & tax | 57.4 | 56.7 | −0.8 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 62.2 | 62.9 | +0.7 |
| AA-AnalystAgent (pass^5) | Professional › General knowledge work | 53.8 | 53.8 | 0.0 |
Data: Artificial Analysis, Vals AI, Arena and 9 othersatlas.village.dev