Muse Wonder Full vs Qwen3.8 2.4T A95B
Full profiles: Muse Wonder Full (Meta) and Qwen3.8 2.4T A95B (Alibaba): every benchmark, price and rank.
Updated · Data from 5 sources
As of 24 Oct 2026, Muse Wonder Full (max) and Qwen3.8 2.4T A95B are neck and neck on quality (70.7 vs 63.1, inside the ±10.0 margin of error) and close on price, so pick on the category that matters to you. Muse Wonder Full leads in every category we measure both in.
Muse Wonder Full benchmarks max Meta
Qwen3.8 2.4T A95B benchmarks AlibabaOpen in the comparison toolWhich to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Reasoning | Muse Wonder Full | 95 vs 74 category score; CritPt 32.9% vs 20.0% | High |
| Knowledge | Muse Wonder Full | 71 vs 63 category score; AA-Omniscience 30.1 vs 4.3 | High |
| Long documents | Muse Wonder Full | 67 vs 65 category score; AA-LCR 83.7% vs 80.3% | Medium |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Muse Wonder Full (max) and Qwen3.8 2.4T A95B are neck and neck on quality (70.7 vs 63.1, within the ±10.0 margin).
- Muse Wonder Full (max) is stronger in coding, composite, knowledge, long context, reasoning.
- Qwen3.8 2.4T A95B is stronger in agents & tools.
At a glance
| Spec | Muse Wonder Full | Qwen3.8 2.4T A95B |
|---|---|---|
| Developer | Meta | Alibaba |
| Released | 12 Oct 2026 | 12 Aug 2026 |
| Context window (tokens) | 1,000,000 | 1,000,000 |
| Compared at | Muse Wonder Full (max) | Qwen3.8 2.4T A95B |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | – | $2.00 / $6.00 / $0.250 |
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 | Muse Wonder Full (max) | Qwen3.8 2.4T A95B | Leader |
|---|---|---|---|
| Reasoning | 94.9 | 73.7 | Muse Wonder Full |
| Coding | 72.3 | – | not comparable |
| Agents & tools | – | 78.4 | not comparable |
| Knowledge | 71.4 | 63.4 | Muse Wonder Full |
| Long context | 67.0 | 65.2 | Muse Wonder Full |
| Composite | 91.9 | 74.2 | Muse Wonder Full |
Where sources disagree
Across the 11 benchmarks both have, Muse Wonder Full leads 11, Qwen3.8 2.4T A95B leads 0. Every overall index that rates both puts Muse Wonder Full ahead.
| Overall index | Muse Wonder Full (max) | Qwen3.8 2.4T A95B | Leader |
|---|---|---|---|
| BenchLeader Index | 70.7 | 63.1 | neck and neck |
| AA Intelligence Index v4.3.2 | 55.8 | 40.0 | Muse Wonder Full |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
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 | Muse Wonder Full (max) | Qwen3.8 2.4T A95B |
|---|---|---|
| Blended price per 1M tokens | – | $3.00 |
| Input / output per 1M | – / – | $2.00 / $6.00 |
| Cached input per 1M | – | $0.250 |
| Price per task: chat reply | – | $0.0026 |
| Price per task: summarize a 30-page report | – | $0.028 |
| Price per task: code edit | – | $0.021 |
| Price per task: agentic coding session | – | $0.144 |
| Price per task: structured extraction | – | $0.0052 |
| Cost to run a full benchmark suite | – | $2.16 |
| Output speed | – | 43 tok/s |
| Time to first answer | – | 49.5 s |
| Time to a complete answer | – | 61.1 s |
Each model is compared at its best-ranked reasoning-effort setting. Muse Wonder Full (max) scores 70.7. Qwen3.8 2.4T A95B scores 63.1 and takes 49 s to its first answer token. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
55 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Muse Wonder Full (max) | Qwen3.8 2.4T A95B |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 70.7 | 63.1 |
| Coding score | BenchLeader | 72.3 | – |
| Composite score | BenchLeader | 91.9 | 74.2 |
| Knowledge score | BenchLeader | 71.4 | 63.4 |
| Long context score | BenchLeader | 67.0 | 65.2 |
| Reasoning score | BenchLeader | 94.9 | 73.7 |
| Agents & tools score | BenchLeader | – | 78.4 |
| Blended price $/M | BenchLeader | – | $3.00 |
| Output speed | BenchLeader | – | 43 tok/s |
| Time to first answer | BenchLeader | – | 49.5 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,000,000 tokens |
| Reasoning | |||
| GPQA Diamond (AA) | Artificial Analysis | – | 93.5% |
| Humanity's Last Exam (AA) | Artificial Analysis | 53.6% | 42.5% |
| CritPt | Artificial Analysis | 32.9% | 20.0% |
| MysteryMechanism | Vals AI | 35.1% | – |
| Coding | |||
| Arena WebDev | Arena (LMArena) | 1726 | – |
| SciCode (AA) | Artificial Analysis | 62.6% | 54.0% |
| IOI | Vals AI | 79.8% | – |
| Code Migration | Vals AI | 61.0% | – |
| Code Migration v1.1 | Vals AI | 62.8% | – |
| ProgramBench | Vals AI | 0.5% | – |
| SWE-bench Multimodal | Vals AI | 47.3% | – |
| Vibe Code Bench 1-100 | Vals AI | 34.0% | – |
| Vibe Code Bench v1.1 | Vals AI | 86.0% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 57.1% | 11.1% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 82.0% |
| FrontierCode 1.1 (AA) | Artificial Analysis | 44.9% | – |
| Agents & tools | |||
| MCP Atlas | Scale AI SEAL | – | 84.5%† |
| GDPval-AA v2.1 | Artificial Analysis | 67.8% | 55.4% |
| τ³-Banking (AA) | Artificial Analysis | – | 49.1% |
| CyberBench | Vals AI | 73.6% | – |
| Finance Agent v2 | Vals AI | 59.1% | – |
| Finance Agent v2.1 | Vals AI | 67.6% | – |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 23.3% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 21.7% | – |
| Legal Research Bench | Vals AI | 53.4% | – |
Show all 55 benchmarks
| Metric | Publisher | Muse Wonder Full (max) | Qwen3.8 2.4T A95B |
|---|---|---|---|
| Agents & tools | |||
| Robo-bench | Vals AI | 10.4% | – |
| SREBench | Vals AI | 11.8% | – |
| Tax Agent Bench | Vals AI | 43.0% | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 58.1% | – |
| Terminal-Bench Science | Vals AI | 37.1% | – |
| AutomationBench | Artificial Analysis | 71.3% | 57.3% |
| GDP.pdf (AA) | Artificial Analysis | 32.6% | 15.6% |
| Harvey LAB-AA v1.1 | Harvey | 13.1% | – |
| EnterpriseOps-Gym | Artificial Analysis | – | 47.4% |
| AA-Briefcase v1.1 | Artificial Analysis | 1729 | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 37.1% | – |
| Knowledge | |||
| AA-Omniscience | Artificial Analysis | 30.1 | 4.3 |
| LegalBench | Vals AI | 87.4% | – |
| BioMysteryBench | Vals AI | 73.3% | – |
| Excel Modeling Benchmark | Vals AI | 71.9% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 75.0% | – |
| MedCode | Vals AI | 49.9% | – |
| MedScribe | Vals AI | 89.3% | – |
| Public Benefits Bench v1.1 | Vals AI | 73.8% | – |
| AA-Omniscience: accuracy | Artificial Analysis | 43.2% | 31.3% |
| AA-Omniscience: non-hallucination | Artificial Analysis | 77.0% | 60.8% |
| Long context | |||
| AA-LCR | Artificial Analysis | 83.7% | 80.3% |
| MLCR | Artificial Analysis | – | 30.6% |
| Composite | |||
| AA Intelligence Index v4.3.2 | Artificial Analysis | 55.8 | 40.0 |
| Vals Index | Vals AI | 66.8 | – |
| Vals Finance Index | Vals AI | 64.5% | – |
| Vals RSI Index | Vals AI | 24.4% | – |
| Vals Science Index | Vals AI | 48.5% | – |
| Vals Router Index | Vals AI | 66.8%† | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | – | 21.6%† |
Data as of 2026-10-24. Best configuration of each model; every score links to its source on the model pages.
What changed
- Muse Wonder Full (max): 1 new result (Terminal-Bench-Science 0.1 (AA)).
- Muse Wonder Full (max): 9 new results (Code Migration v1.1, Excel Modeling Benchmark v1.1, Finance Agent v2.1…).
- Muse Wonder Full (max): 1 new result (FrontierCode 1.1 (AA)).
- Qwen3.8 2.4T A95B: 1 score revision.
- Muse Wonder Full (max): BenchLeader Index 68.6 → 70.7.
- Muse Wonder Full (max): 1 new result (Arena WebDev); 2 score revisions.
- Qwen3.8 2.4T A95B: 4 new results (AutomationBench, GDP.pdf (AA), MLCR…).
- Muse Wonder Full (max): 21 new results (IOI, LegalBench, Vals Index…).
- Muse Wonder Full (max): 1 new result (Harvey LAB-AA v1.1).
- Muse Wonder Full (max): first results.
Related comparisons
- Muse Wonder Full vs Claude Opus 5.5
- Muse Wonder Full vs Gemini 4 Argon
- Muse Wonder Full vs Claude Fable 5.1
- Qwen3.8 2.4T A95B vs Claude Opus 5.5
- Qwen3.8 2.4T A95B vs Gemini 4 Argon
- Qwen3.8 2.4T A95B vs Claude Fable 5.1
- Muse Spark 1.3 vs Qwen3.8 2.4T A95B
- Muse Wonder Full vs Qwen3.8 Max
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Muse Wonder Full vs Qwen3.8 2.4T A95B: questions
- Is Muse Wonder Full better than Qwen3.8 2.4T A95B?
- Muse Wonder Full (max) and Qwen3.8 2.4T A95B are neck and neck on quality (70.7 vs 63.1, within the ±10.0 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-24, so check the category scores for your use.
- Which is better for long documents?
- Muse Wonder Full scores higher on long-context benchmarks (67 vs 65). Both accept 1,000,000 tokens of context.
- Do independent benchmarks agree on Muse Wonder Full vs Qwen3.8 2.4T A95B?
- Across the 11 benchmarks both have, Muse Wonder Full leads 11, Qwen3.8 2.4T A95B leads 0. Every overall index that rates both puts Muse Wonder Full ahead.
- When was Muse Wonder Full released?
- Muse Wonder Full was released on 12 Oct 2026.