GPT-5.6 Sol vs Muse Wonder Full
As of 18 Oct 2026, Muse Wonder Full (max) is the better pick for most workloads: it leads on quality (68.6 vs 67.3 on the BenchLeader Index).
GPT-5.6 Sol 
Muse Wonder Full Key differences
- Muse Wonder Full (max) leads on quality: 68.6 vs 67.3.
- GPT-5.6 Sol (max) is stronger in agents & tools, coding, instruction following, long context, maths, multimodal.
- Muse Wonder Full (max) is stronger in composite, knowledge, reasoning.
Choose GPT-5.6 Sol (max) if
- coding is the priority (66 vs 64 on the category score)
Choose Muse Wonder Full (max) if
- reasoning is the priority (95 vs 72 on the category score)
- knowledge is the priority (71 vs 65 on the category score)
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | GPT-5.6 Sol (max) | Muse Wonder Full (max) | Leader |
|---|---|---|---|
| Reasoning | 72.4 | 95.0 | Muse Wonder Full |
| Coding | 65.9 | 64.0 | GPT-5.6 Sol |
| Agents & tools | 66.4 | – | not comparable |
| Maths | 71.1 | – | not comparable |
| Knowledge | 65.4 | 71.3 | Muse Wonder Full |
| Instruction following | 71.0 | – | not comparable |
| Multimodal | 66.6 | – | not comparable |
| Long context | 67.1 | 67.0 | level |
| Composite | 74.9 | 92.2 | Muse Wonder Full |
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | GPT-5.6 Sol (max) | Muse Wonder Full (max) |
|---|---|---|
| Blended price per 1M tokens | $8.00 | – |
| Input / output per 1M | $4.00 / $20.00 | – / – |
| Price per task: chat reply | $0.0076 | – |
| Price per task: summarise a 30-page report | $0.060 | – |
| Price per task: code edit | $0.054 | – |
| Price per task: agentic coding session | $0.320 | – |
| Price per task: structured extraction | $0.012 | – |
| Cost to run a full benchmark suite | $1.99 | – |
| Output speed | 89 tok/s | – |
| Time to first answer | 118.9 s | – |
| Time to a complete answer | 124.5 s | – |
Every benchmark
| Metric | GPT-5.6 Sol (max) | Muse Wonder Full (max) |
|---|---|---|
| BenchLeader Index | 67.3 | 68.6 |
| Agents & tools score | 66.4 | – |
| Coding score | 65.9 | 64.0 |
| Composite score | 74.9 | 92.2 |
| Instruction following score | 71.0 | – |
| Knowledge score | 65.4 | 71.3 |
| Long context score | 67.1 | 67.0 |
| Maths score | 71.1 | – |
| Multimodal score | 66.6 | – |
| Reasoning score | 72.4 | 95.0 |
| Blended price $/M | $8.00 | – |
| Output speed | 89 tok/s | – |
| Time to first answer | 118.9 s | – |
| Context window | 1.1M | 1M |
| GPQA Diamond | 93.5% | – |
| FrontierMath Tiers 1–3 | 89.1% | – |
| FrontierMath Tier 4 | 82.9% | – |
| OTIS Mock AIME | 100.0% | – |
| SimpleQA Verified | 69.7% | – |
| Terminal-Bench | 37.3% | – |
| OSWorld-Verified 2.0 | 27.3% | – |
| SciCode | 57.1% | – |
| WeirdML | 87.0% | – |
| ProofBench | 83.0% | – |
| LiveBench | 81.0% | – |
| LiveBench Reasoning | 91.7% | – |
| LiveBench Coding | 83.9% | – |
| LiveBench Agentic Coding | 56.2% | – |
| LiveBench Mathematics | 96.2% | – |
| LiveBench Data Analysis | 79.8% | – |
| LiveBench Language | 87.7% | – |
| LiveBench Instruction Following | 71.8% | – |
| AA Intelligence Index v4.3.2 | 47.0 | 55.8 |
| IFBench | 72.7% | – |
| AA-LCR | 84.0% | 83.7% |
| MMMU-Pro | 83.4% | – |
| AA-Omniscience | 22.0 | 30.1 |
| Terminal-Bench Hard | 65.9% | – |
| GPQA Diamond (AA) | 94.1% | – |
| Humanity's Last Exam (AA) | 49.5% | 53.6% |
| SciCode (AA) | 57.1% | 62.6% |
| τ²-Bench Telecom (AA) | 85.1% | – |
| LiveCodeBench | 82.6% | – |
| MMLU-Pro | 89.1% | – |
| IOI | 91.2% | 79.8% |
| LegalBench | 87.0% | 87.4% |
| CorpFin | 64.4% | – |
| TaxEval | 74.8% | – |
| Terminal-Bench 2.1 (Vals) | 85.8% | – |
| SWE-bench (Vals) | 96.2% | – |
| GPQA Diamond (Vals) | 95.2% | – |
| Vals Index | 58.0 | 64.6 |
| PRBench Finance | 50.5% | – |
| PRBench Legal | 50.5% | – |
| ARC-AGI-1 | 96.5% | – |
| ARC-AGI-2 | 92.5% | – |
| ARC-AGI-3 | 7.8% | – |
| CritPt | 32.3% | 32.9% |
| GDPval-AA v2.1 | 55.4% | 67.8% |
| τ³-Banking (AA) | 44.3% | – |
| ITBench SRE (AA) | 56.2% | – |
| Analyst Agent (AA) | 47.5% | – |
| BioMysteryBench | 71.1% | 73.3% |
| Code Migration | 52.9% | 61.0% |
| CUA-bench | 8.3% | – |
| CyberBench | 76.3% | 73.6% |
| Excel Modeling Benchmark | 72.3% | 71.9% |
| Finance Agent v2 | 53.8% | 59.1% |
| Harvey's Legal Agent Benchmark | 2.5% | 21.7% |
| Legal Research Bench | 48.1% | 53.4% |
| MedCode | 44.0% | 49.9% |
| MedScribe | 85.2% | 89.3% |
| MMMU-Pro (Vals) | 88.8% | – |
| MortgageTax | 67.3% | – |
| MysteryMechanism | 33.3% | 35.1% |
| ProgramBench | 1.5% | 0.5% |
| Public Benefits Bench | 66.5% | 73.8% |
| SAGE | 52.6% | – |
| SkillsBench | 54.1% | – |
| SREBench | 30.5% | 11.8% |
| Tax Agent Bench | 31.1% | 43.0% |
| Terminal-Bench 4.0 (Vals) | 37.9% | 58.1% |
| Terminal-Bench Science | 20.0% | 37.1% |
| Time Horizon Index: KSP | 23.8% | – |
| Vals Multimodal Index | 72.6% | – |
| Vibe Code Bench 1-100 | 20.0% | 34.0% |
| Vibe Code Bench v1.1 | 80.5% | 86.0% |
| Web Search Index | 43.6% | – |
| DrugDiscoveryBench | 56.5% | – |
| Chess Puzzles | 59.0% | – |
| EBR-bench | 44.8% | – |
| Mystery Game Puzzles | 58.0% | – |
| BALROG | 60.0% | – |
| PostTrainBench | 36.2% | – |
| DeepSWE v1.1 | 72.7% | – |
| LMCA | 58.4% | – |
| DTBench | 95.5% | – |
| CursorBench | 41.7% | – |
| ALE-Bench | 2176.9 | – |
| GDP.pdf | 30.7% | – |
| FrontierSWE | 32.2% | – |
| BTF-3 | 13.7% | – |
| Terminal-Bench 4.0 (AA) | 39.9% | 57.1% |
| Terminal-Bench 2.1 (AA) | 88.0% | – |
| AutomationBench | 60.1% | 71.3% |
| GDP.pdf | 27.2% | 32.6% |
| MLCR | 26.1% | – |
| Harvey LAB-AA v1.1 | – | 13.1% |
| EnterpriseOps-Gym | 42.9% | – |
| AA-Omniscience: accuracy | 59.4% | 43.2% |
| AA-Omniscience: non-hallucination | 7.8% | 77.0% |
| AA-Briefcase v1.1 | – | 1729 |
| Terminal-Bench-Science 0.1 (AA) | 22.4% | – |
Data as of 2026-10-18. Best configuration of each model; every score links to its source on the model pages.
GPT-5.6 Sol vs Muse Wonder Full: questions
- Is GPT-5.6 Sol better than Muse Wonder Full?
- Muse Wonder Full (max) leads on quality: 68.6 vs 67.3. The BenchLeader Index combines every independent quality benchmark; Muse Wonder Full (max) is ahead overall as of 2026-10-18, but check the category scores for your use.
- Is GPT-5.6 Sol better than Muse Wonder Full for coding?
- GPT-5.6 Sol scores higher in coding (66 vs 64 on the category index, where 50 is average).
- Which has the larger context window?
- GPT-5.6 Sol accepts more context: 1.1M against 1M tokens.