Claude Opus 4.6 vs Muse Spark
Full profiles: Claude Opus 4.6 (Anthropic) and Muse Spark (Meta): every benchmark, price and rank.
Updated · Data from 10 sources
As of 25 Oct 2026, Claude Opus 4.6 (high) and Muse Spark are neck and neck on quality (62.7 vs 64.3, inside the ±3.9 margin of error) and close on price, so pick on the category that matters to you. Pick Claude Opus 4.6 for maths (64 vs 59); pick Muse Spark for coding (65 vs 63).
Claude Opus 4.6 benchmarks 
Muse Spark benchmarks Which 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 |
|---|---|---|---|
| Coding | Muse Spark | 65 vs 63 category score; Arena Coding 1529 vs 1547 | Medium |
| Maths | Claude Opus 4.6 | 64 vs 59 category score; Arena Maths 1517 vs 1467 | Medium |
| Reasoning | Either | 66 vs 65 category score; Arena Hard Prompts 1530 vs 1506 | Either |
| Images and other media | Claude Opus 4.6 | 67 vs 65 category score; Arena Vision 1316 vs 1306 | Medium |
| Chat quality (human preference) | Claude Opus 4.6 | 70 vs 68 on Arena votes | Medium |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Claude Opus 4.6 (high) and Muse Spark are neck and neck on quality (62.7 vs 64.3, within the ±3.9 margin).
- Claude Opus 4.6 (high) is stronger in human preference, maths, multimodal, reasoning.
- Muse Spark is stronger in coding, agents & tools, composite, instruction following, knowledge, long context.
At a glance
| Spec | Claude Opus 4.6 | Muse Spark |
|---|---|---|
| Developer | Anthropic | Meta |
| Released | 5 Feb 2026 | 8 Apr 2026 |
| Context window (tokens) | 1,000,000 | 262,144 |
| Compared at | Claude Opus 4.6 (high) | Muse Spark |
| Open weights | No | No |
| Price in / out / cached input per 1M | $5.00 / $25.00 / $0.500 | – |
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 | Claude Opus 4.6 (high) | Muse Spark | Leader |
|---|---|---|---|
| Reasoning | 65.5 | 65.3 | level |
| Coding | 62.7 | 65.0 | Muse Spark |
| Agents & tools | – | 66.1 | not comparable |
| Maths | 63.5 | 59.1 | Claude Opus 4.6 |
| Knowledge | – | 63.3 | not comparable |
| Instruction following | – | 77.3 | not comparable |
| Human preference | 70.0 | 68.3 | Claude Opus 4.6 |
| Multimodal | 66.6 | 64.5 | Claude Opus 4.6 |
| Long context | – | 64.0 | not comparable |
| Composite | – | 64.5 | not comparable |
Where sources disagree
Across the 9 benchmarks both have, Claude Opus 4.6 leads 9, Muse Spark leads 0. Independent sources split: Claude Opus 4.6 leads on Arena Text; Muse Spark leads on AA Intelligence Index v4.3.2.
| Overall index | Claude Opus 4.6 (high) | Muse Spark | Leader |
|---|---|---|---|
| BenchLeader Index | 62.7 | 64.3 | neck and neck |
| AA Intelligence Index v4.3.2 | 26.4† | 31.3 | Muse Spark |
| Arena Text | 1503 | 1489 | Claude Opus 4.6 |
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 | Claude Opus 4.6 (high) | Muse Spark |
|---|---|---|
| Blended price per 1M tokens | $10.00 | – |
| Input / output per 1M | $5.00 / $25.00 | – / – |
| Cached input per 1M | $0.500 | – |
| Price per task: chat reply | $0.0095 | – |
| Price per task: summarize a 30-page report | $0.075 | – |
| Price per task: code edit | $0.068 | – |
| Price per task: agentic coding session | $0.400 | – |
| Price per task: structured extraction | $0.015 | – |
| Output speed | 39 tok/s | – |
| Time to first answer | 1.2 s | – |
Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 4.6 (high) scores 62.7 and takes 1 s to its first answer token; Claude Opus 4.6 (thinking) scores 62.5 in 1 s, Claude Opus 4.6 scores 62.5 in 1 s. Muse Spark scores 64.3. 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
113 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 | Claude Opus 4.6 (high) | Muse Spark |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 62.7 | 64.3 |
| Coding score | BenchLeader | 62.7 | 65.0 |
| Human preference score | BenchLeader | 70.0 | 68.3 |
| Maths score | BenchLeader | 63.5 | 59.1 |
| Multimodal score | BenchLeader | 66.6 | 64.5 |
| Reasoning score | BenchLeader | 65.5 | 65.3 |
| Agents & tools score | BenchLeader | – | 66.1 |
| Composite score | BenchLeader | – | 64.5 |
| Instruction following score | BenchLeader | – | 77.3 |
| Knowledge score | BenchLeader | – | 63.3 |
| Long context score | BenchLeader | – | 64.0 |
| Blended price $/M | BenchLeader | $10.00 | – |
| Output speed | BenchLeader | 39 tok/s | – |
| Time to first answer | BenchLeader | 1.2 s | – |
| Context window | BenchLeader | 1,000,000 tokens | 262,144 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 90.5%† | 89.8% |
| Humanity's Last Exam | Scale AI / CAIS | 19.0%† | 40.6% |
| SimpleBench | SimpleBench | 67.6%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1530 | 1506 |
| LiveBench 1.0 Reasoning | LiveBench | 88.7% | – |
| GPQA Diamond (AA) | Artificial Analysis | 89.6%† | 88.4% |
| Humanity's Last Exam (AA) | Artificial Analysis | 39.9%† | 40.7% |
| GPQA Diamond (Vals) | Vals AI | 89.7%† | 89.7% |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 72.4%† | – |
| ARC-AGI-1 | ARC Prize | 94.0% | – |
| ARC-AGI-2 | ARC Prize | 69.2% | – |
| ARC-AGI-3 | ARC Prize | 0.5%† | – |
| CritPt | Artificial Analysis | 12.6%† | 11.3% |
| Chess Puzzles | Epoch AI Benchmarking Hub | 21.0%† | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 15.0%† | – |
| CL-bench | Epoch AI Benchmarking Hub | 20.7%† | – |
| CL-bench Life | Epoch AI Benchmarking Hub | 17.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 55.8%† | – |
| DTBench | Epoch AI Benchmarking Hub | 91.2%† | – |
| ForecastBench | ForecastBench | 63.4%† | – |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 78.7%† | – |
| SciCode | SciCode | – | 51.5% |
| WeirdML | WeirdML | 78.0% | – |
| FrontierCode | Cognition | 26.6%† | – |
| GSO-Bench | GSO-Bench | 41.2% | – |
Show all 113 benchmarks
| Metric | Publisher | Claude Opus 4.6 (high) | Muse Spark |
|---|---|---|---|
| Coding | |||
| Arena Coding | Arena (LMArena) | 1547 | 1529 |
| Arena WebDev | Arena (LMArena) | 1546 | – |
| LiveBench 1.0 Coding | LiveBench | 78.2% | – |
| LiveCodeBench | Vals AI | 84.7%† | – |
| SWE-bench (Vals) | Vals AI | 78.2%† | 74.4% |
| SWE-Bench Pro | Scale AI SEAL | 51.9%† | 55.0% |
| SWE-bench Verified (bash only) | SWE-bench | 75.6%† | – |
| SWE-bench Verified (any scaffold) | SWE-bench | 75.6%† | – |
| Vibe Code Bench v1.1 | Vals AI | 57.6%† | 19.7% |
| SWE-Bench Pro (private) | Scale AI SEAL | 51.9%† | 55.0% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 33.3%† | 24.2% |
| SWE Atlas: Refactoring | Scale AI SEAL | 35.6%† | – |
| SWE Atlas: Test Writing | Scale AI SEAL | 36.7%† | 31.1% |
| ALE-Bench | Epoch AI Benchmarking Hub | 996.5† | – |
| AlgoTune | AlgoTune | 1.5 | – |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 62.2% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 79.8%† | – |
| Cybench | Cybench | 93.0%† | – |
| Remote Labor Index | Scale AI / CAIS | 4.2%† | – |
| APEX-Agents | Mercor | 46.3%† | – |
| LiveBench Agentic Coding | LiveBench | 49.0% | – |
| Terminal-Bench Hard | Artificial Analysis | 46.2%† | 45.5% |
| τ²-Bench Telecom (AA) | Artificial Analysis | 92.1%† | 91.5% |
| MCP Atlas | Scale AI SEAL | 76.8%† | 82.2% |
| HiL-Bench | Scale AI SEAL | 38.3%† | – |
| τ³-bench Banking | τ²-bench | 27.3%† | – |
| APEX-Agents (AA) | Artificial Analysis | 33.0%† | – |
| Terminal-Bench 2.0 (Vals) | Vals AI | 58.4%† | 59.5% |
| DeepResearch Bench | DeepResearch Bench | 55.3% | – |
| METR Time Horizons | METR | 78.9%† | – |
| GBAEval | Epoch AI Benchmarking Hub | 44.1%† | – |
| Vending-Bench 2 | Andon Labs | 8017.6† | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 66.0%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 26.8%† | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 94.4%† | 88.9% |
| ProofBench | Vals AI | 50.0%† | 17.0% |
| LiveBench Mathematics | LiveBench | 89.3% | – |
| AIME (Vals) | Vals AI | 95.6%† | 96.9% |
| AIME 2026 | MathArena | 96.7% | – |
| HMMT February 2026 | MathArena | 96.2% | – |
| USAMO 2026 | MathArena | 47.0% | – |
| MathArena Apex | MathArena | 34.5% | – |
| Arena Maths | Arena (LMArena) | 1517 | 1467 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 47.0%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 69.9% | – |
| AA-Omniscience | Artificial Analysis | 13.7† | 7.2 |
| MMLU-Pro | Vals AI | 89.1%† | 87.3% |
| LegalBench | Vals AI | 85.3%† | 84.2% |
| CorpFin | Vals AI | 67.0%† | 65.1% |
| TaxEval | Vals AI | 76.0%† | 77.7% |
| MedQA | Vals AI | 95.4%† | – |
| PRBench Finance | Scale AI SEAL | 53.3%† | 52.4% |
| PRBench Legal | Scale AI SEAL | 52.3%† | 52.3% |
| MultiNRC | Scale AI SEAL | 57.1%† | 59.0% |
| CaseLaw v2 | Vals AI | 62.1%† | 63.1% |
| MedCode | Vals AI | 48.2%† | 51.3% |
| MedScribe | Vals AI | 86.7%† | 85.9% |
| DrugDiscoveryBench | Scale AI SEAL | 27.7%† | – |
| EBR-bench | Epoch AI Benchmarking Hub | 12.7%† | – |
| AA-Omniscience: accuracy | Artificial Analysis | 47.0%† | 49.6% |
| AA-Omniscience: non-hallucination | Artificial Analysis | 37.2%† | 15.8% |
| Global-MMLU-Lite (AA) | Artificial Analysis | 92.2%† | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 83.3% | – |
| LiveBench Instruction Following | LiveBench | 63.3% | – |
| IFBench | Artificial Analysis | 53.1%† | 75.9% |
| MultiChallenge | Scale AI SEAL | 37.1%† | 75.5% |
| TutorBench | Scale AI SEAL | 53.7%† | 68.5% |
| Arena Instruction Following | Arena (LMArena) | 1511 | 1465 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1503 | 1489 |
| EQ-Bench 4 | EQ-Bench | 1213† | – |
| Arena Creative Writing | Arena (LMArena) | 1500 | 1464 |
| Arena Multi-turn | Arena (LMArena) | 1516 | 1494 |
| Arena Longer Queries | Arena (LMArena) | 1521 | 1476 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1316 | 1306 |
| MMMU-Pro | Artificial Analysis | 75.4%† | 80.5% |
| VISTA | Scale AI SEAL | 46.1%† | – |
| MMMU-Pro (official) | MMMU | – | 80.4%† |
| MMMU-Pro (Vals) | Vals AI | 83.9%† | 87.4% |
| MortgageTax | Vals AI | 68.5%† | – |
| SAGE | Vals AI | 51.6%† | – |
| VTB | Scale AI SEAL | 27.5%† | – |
| Long context | |||
| AA-LCR | Artificial Analysis | 78.0%† | 78.0% |
| Arena Document | Arena (LMArena) | 1495 | 1467 |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 155.7† | 152.0 |
| LiveBench 1.0 | LiveBench | 74.5% | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 31.9† | 31.3 |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 13.0%† | 20.2% |
| MASK | Scale AI SEAL | 85.4%† | – |
Data as of 2026-10-25. Best configuration of each model; every score links to its source on the model pages.
What changed
- Claude Opus 4.6 (high): BenchLeader Index 60.4 → 62.9.
- Claude Opus 4.6 (high): 1 new result (USAMO 2026).
- Muse Spark: 1 new result (PRBench Legal).
- Muse Spark: 1 new result (SWE Atlas: Codebase QnA).
- Muse Spark: 2 new results (PRBench Legal, MultiNRC).
- Muse Spark: 1 new result (PRBench Finance).
- Muse Spark: 1 new result (PRBench Finance).
- Muse Spark: 1 new result (TutorBench); 1 score revision.
- Muse Spark: BenchLeader Index 64.9 → 64.4.
- Muse Spark: 1 new result (PRBench Finance).
Related comparisons
- Claude Opus 4.6 vs Claude Opus 5.5
- Claude Opus 4.6 vs Gemini 4 Argon
- Claude Opus 4.6 vs Muse Wonder Full
- Muse Spark vs Claude Opus 5.5
- Muse Spark vs Gemini 4 Argon
- Muse Spark vs Muse Wonder Full
- Muse Spark vs Claude Fable 5.1
- Claude Opus 4.6 vs GPT-6 Astra
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Claude Opus 4.6 vs Muse Spark: questions
- Is Claude Opus 4.6 better than Muse Spark?
- Claude Opus 4.6 (high) and Muse Spark are neck and neck on quality (62.7 vs 64.3, within the ±3.9 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-25, so check the category scores for your use.
- Is Claude Opus 4.6 better than Muse Spark for coding?
- Muse Spark scores higher in coding (65 vs 63 on the category index, where 50 is average). On Arena Coding, the top shared coding benchmark, it is 1547 for Claude Opus 4.6 against 1529 for Muse Spark.
- Which is better for long documents?
- Claude Opus 4.6 accepts more context: 1,000,000 against 262,144 tokens for Muse Spark.
- Do independent benchmarks agree on Claude Opus 4.6 vs Muse Spark?
- Across the 9 benchmarks both have, Claude Opus 4.6 leads 9, Muse Spark leads 0. Independent sources split: Claude Opus 4.6 leads on Arena Text; Muse Spark leads on AA Intelligence Index v4.3.2.
- When was Muse Spark released?
- Muse Spark was released on 8 Apr 2026.