GPT-6.1 Sol vs Muse Ember
GPT-6.1 Sol leads by 4.0 capability points (≈95% interval +1.5 to +6.5): a clear lead over Muse Ember · P(higher) ≈ 100%
Wins 33 of 53 shared benchmarks (20 lost)
Capability vs. relative cost
GPT-6.1 SolMuse EmberClosed weightsOpen weights
Benchmark wins33 – 20 · 53 shared
Comparing matching setups only (11 benchmarks excluded) · suites counted once
Area← GPT-6.1 SolMuse Ember →
Data: Artificial Analysis, Vals AI, Surge AI and 10 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Ember minus GPT-6.1 Sol.
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 58.1 | 20.0 | −38.1 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 36.0 | 2.0 | −34.0 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 58.8 | 25.1 | −33.7 |
| DAXBench v2 | Software › Data & ML | 92.6 | 62.8 | −29.8 |
| KradleBench | Games | 54.6 | 25.7 | −28.9 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 75.1 | 47.1 | −28.0 |
| RiemannBench (November 2026 rescoring) | Math | 79.2 | 53.6 | −25.6 |
| BuseyBench SVG | Perception & media › Generation | 8.0 | 5.4 | −25.5 |
| GBENCH One-shot Coding GScore | Games | 77.5 | 56.4 | −21.1 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 82.6 | 62.3 | −20.3 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 11.3 | 1.0 | −20.2 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 38.3 | 18.3 | −20.0 |
| HLAB-Vals-1.1 | Professional › Legal | 6.3 | 25.0 | +18.8 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 54.6 | 36.0 | −18.6 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 42.4 | 7.7 | −17.3 |
| Terminal-Bench 4.0 | Software › Software engineering | 55.1 | 39.4 | −15.7 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 56.1 | 40.9 | −15.2 |
| Harvey LAB (Mercor run, undisclosed ~213-task cohort) | Professional › Legal | 8.8 | 22.7 | +13.9 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 24.5 | -2.7 | −13.6 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 40.6 | 27.2 | −13.5 |
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 50.3 | 62.1 | +11.8 |
| HiL-Bench (Human-in-Loop Benchmark) | Tool & computer use › General agents | 62.3 | 74.0 | +11.7 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 70.1 | 58.8 | −11.3 |
| Legal Research Bench | Professional › Legal | 38.5 | 49.0 | +10.6 |
| τ³-Banking (Mercor run, bm25_grep retrieval) | Tool & computer use › Customer service & operations | 43.0 | 53.3 | +10.3 |
| SimpleBench | Knowledge & reasoning › General knowledge | 82.9 | 72.7 | −10.2 |
| SWE Atlas v1.1 - Overall | Software › Software engineering | 65.9 | 56.0 | −9.9 |
| Professional Reasoning Benchmark - Finance | Professional › Finance, accounting & tax | 46.2 | 56.0 | +9.8 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 33.5 | 24.2 | −9.3 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 32.0 | 22.8 | −9.2 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 62.3 | 71.4 | +9.0 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 25.2 | 17.2 | −8.0 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 47.9 | 55.6 | +7.7 |
| Vals Legal Index | Professional › Legal | 68.7 | 75.7 | +7.0 |
| Artificial Analysis Progress Index v1 | Composite indexes | 348 | 329 | −6.9 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 11.6 | -2.0 | −6.8 |
| GDP.xlsx | Professional › General knowledge work | 22.0 | 15.4 | −6.6 |
| APEX-SWE | Software › Software engineering | 46.9 | 52.9 | +6.0 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 88.9 | 83.7 | −5.2 |
| Vals Finance Index | Professional › Finance, accounting & tax | 55.8 | 60.9 | +5.1 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 11.8 | 4.8 | −3.5 |
| Vals Index v2.2 | Composite indexes | 63.9 | 60.5 | −3.5 |
| FORTRESS Average Risk Score (ARS) | Safety & alignment | 14.7 | 11.9 | +2.8 |
| APEX-Agents 1.1 | Professional › General knowledge work | 73.2 | 75.7 | +2.5 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 52.0 | 54.4 | +2.4 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 51.6 | 49.2 | −2.4 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 61.8 | 63.8 | +2.0 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1557 | 1520 | −2.0 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 66.6 | 68.1 | +1.5 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 28.6 | 29.8 | +1.2 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 71.1 | 72.2 | +1.1 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | 5.1 | 6.6 | +0.8 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.2 | 0.9 | −0.1 |
Data: Artificial Analysis, Vals AI, Surge AI and 10 othersatlas.village.dev