Muse Spark vs Muse Spark 1.1
Full profiles: Muse Spark (Meta) and Muse Spark 1.1 (Meta): every benchmark, price and rank.
Updated · Data from 8 sources
As of 25 Oct 2026, Muse Spark and Muse Spark 1.1 are neck and neck on quality (64.3 vs 62.5, inside the ±6.4 margin of error) and close on price, so pick on the category that matters to you. Pick Muse Spark for agents & tools (66 vs 52); pick Muse Spark 1.1 for knowledge (69 vs 63).
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 1.1 | 66 vs 65 category score; SWE-Bench Pro 61.5% vs 55.0% | Medium |
| Agents & tool use | Muse Spark | 66 vs 52 category score; MCP Atlas 82.2% vs 88.1% | High |
| Maths | Muse Spark 1.1 | 61 vs 59 category score; Arena Maths 1493 vs 1467 | Medium |
| Reasoning | Muse Spark 1.1 | 68 vs 65 category score; Arena Hard Prompts 1510 vs 1506 | Medium |
| Knowledge | Muse Spark 1.1 | 69 vs 63 category score; PRBench Finance 55.0% vs 52.4% | Medium |
| Images and other media | Either | 65 vs 64 category score; Arena Vision 1306 vs 1293 | Either |
| Following instructions | Muse Spark | 77 vs 72 category score; MultiChallenge 75.5% vs 75.3% | Medium |
| Chat quality (human preference) | Muse Spark | 68 vs 65 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
- Muse Spark and Muse Spark 1.1 are neck and neck on quality (64.3 vs 62.5, within the ±6.4 margin).
- Muse Spark is stronger in agents & tools, composite, human preference, instruction following, long context, multimodal.
- Muse Spark 1.1 is stronger in coding, knowledge, maths, reasoning.
At a glance
| Spec | Muse Spark | Muse Spark 1.1 |
|---|---|---|
| Developer | Meta | Meta |
| Released | 8 Apr 2026 | 9 Jul 2026 |
| Context window (tokens) | 262,144 | 1,048,576 |
| Compared at | Muse Spark | Muse Spark 1.1 |
| Open weights | No | No |
| Price in / out / cached input per 1M | – | $1.25 / $4.25 / $0.150 |
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 Spark | Muse Spark 1.1 | Leader |
|---|---|---|---|
| Reasoning | 65.3 | 67.9 | Muse Spark 1.1 |
| Coding | 65.0 | 66.3 | Muse Spark 1.1 |
| Agents & tools | 66.1 | 51.8 | Muse Spark |
| Maths | 59.1 | 61.3 | Muse Spark 1.1 |
| Knowledge | 63.3 | 69.2 | Muse Spark 1.1 |
| Instruction following | 77.3 | 71.5 | Muse Spark |
| Human preference | 68.3 | 65.0 | Muse Spark |
| Multimodal | 64.5 | 63.6 | level |
| Long context | 64.0 | – | not comparable |
| Composite | 64.5 | – | not comparable |
Where sources disagree
Across the 20 benchmarks both have, Muse Spark leads 4, Muse Spark 1.1 leads 16. Every overall index that rates both puts Muse Spark 1.1 ahead.
| Overall index | Muse Spark | Muse Spark 1.1 | Leader |
|---|---|---|---|
| BenchLeader Index | 64.3 | 62.5 | neck and neck |
| AA Intelligence Index v4.3.2 | 31.3 | 33.7† | Muse Spark 1.1 |
| Arena Text | 1489 | 1490 | Muse Spark 1.1 |
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 Spark | Muse Spark 1.1 |
|---|---|---|
| Blended price per 1M tokens | – | $2.00 |
| Input / output per 1M | – / – | $1.25 / $4.25 |
| Cached input per 1M | – | $0.150 |
| Price per task: chat reply | – | $0.0018 |
| Price per task: summarize a 30-page report | – | $0.018 |
| Price per task: code edit | – | $0.014 |
| Price per task: agentic coding session | – | $0.092 |
| Price per task: structured extraction | – | $0.0034 |
| Output speed | – | 200 tok/s |
| Time to first answer | – | 1.5 s |
Each model is compared at its best-ranked reasoning-effort setting. Muse Spark scores 64.3. Muse Spark 1.1 scores 62.5 and takes 2 s to its first answer token; Muse Spark 1.1 (xhigh) scores 61.6 in 18 s. 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
97 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 Spark | Muse Spark 1.1 |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 64.3 | 62.5 |
| Agents & tools score | BenchLeader | 66.1 | 51.8 |
| Coding score | BenchLeader | 65.0 | 66.3 |
| Composite score | BenchLeader | 64.5 | – |
| Human preference score | BenchLeader | 68.3 | 65.0 |
| Instruction following score | BenchLeader | 77.3 | 71.5 |
| Knowledge score | BenchLeader | 63.3 | 69.2 |
| Long context score | BenchLeader | 64.0 | – |
| Maths score | BenchLeader | 59.1 | 61.3 |
| Multimodal score | BenchLeader | 64.5 | 63.6 |
| Reasoning score | BenchLeader | 65.3 | 67.9 |
| Blended price $/M | BenchLeader | – | $2.00 |
| Output speed | BenchLeader | – | 200 tok/s |
| Time to first answer | BenchLeader | – | 1.5 s |
| Context window | BenchLeader | 262,144 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 89.8% | – |
| Humanity's Last Exam | Scale AI / CAIS | 40.6% | – |
| Arena Hard Prompts | Arena (LMArena) | 1506 | 1510 |
| LiveBench 1.0 Reasoning | LiveBench | – | 87.7%† |
| GPQA Diamond (AA) | Artificial Analysis | 88.4% | 89.8%† |
| Humanity's Last Exam (AA) | Artificial Analysis | 40.7% | 46.2%† |
| GPQA Diamond (Vals) | Vals AI | 89.7% | 91.2%† |
| CritPt | Artificial Analysis | 11.3% | 15.1%† |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 34.0%† |
| LMCA | Epoch AI Benchmarking Hub | – | 49.9%† |
| DTBench | Epoch AI Benchmarking Hub | – | 94.4%† |
| Coding | |||
| SciCode | SciCode | 51.5% | 58.8% |
| Arena Coding | Arena (LMArena) | 1529 | 1533 |
| Arena WebDev | Arena (LMArena) | – | 1543 |
| LiveBench 1.0 Coding | LiveBench | – | 77.2%† |
| SciCode (AA) | Artificial Analysis | – | 58.8%† |
| LiveCodeBench | Vals AI | – | 85.9%† |
| SWE-bench (Vals) | Vals AI | 74.4% | 82.0%† |
| SWE-Bench Pro | Scale AI SEAL | 55.0% | 61.5% |
| Code Migration | Vals AI | – | 31.1%† |
| ProgramBench | Vals AI | – | 0.0%† |
| Vibe Code Bench v1.1 | Vals AI | 19.7% | 72.2%† |
| SWE-Bench Pro (private) | Scale AI SEAL | 55.0% | 61.5% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 24.2% | 42.2%† |
| SWE Atlas: Test Writing | Scale AI SEAL | 31.1% | 41.5%† |
Show all 97 benchmarks
| Metric | Publisher | Muse Spark | Muse Spark 1.1 |
|---|---|---|---|
| Coding | |||
| Surface Evolver Bench | Epoch AI Benchmarking Hub | – | 52.5%† |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | – | 53.3% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | – | 6.1%† |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 62.2% | 77.9%† |
| Agents & tools | |||
| APEX-Agents | Mercor | – | 31.8% |
| Arena Agent | Arena (LMArena) | – | -6.9 |
| LiveBench Agentic Coding | LiveBench | – | 58.5%† |
| Terminal-Bench Hard | Artificial Analysis | 45.5% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 91.5% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 69.3%† |
| MCP Atlas | Scale AI SEAL | 82.2% | 88.1% |
| τ³-bench Banking | τ²-bench | – | 40.5%† |
| τ³-Banking (AA) | Artificial Analysis | – | 31.8%† |
| Finance Agent v2 | Vals AI | – | 57.2%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 20.0%† |
| Legal Research Bench | Vals AI | – | 38.0%† |
| SkillsBench | Vals AI | – | 59.2%† |
| Terminal-Bench 2.0 (Vals) | Vals AI | 59.5% | – |
| Time Horizon Index: KSP | Vals AI | – | 1.7%† |
| GBAEval | Epoch AI Benchmarking Hub | – | 7.9% |
| Vending-Bench 2 | Andon Labs | – | 6520.5 |
| AutomationBench | Artificial Analysis | – | 38.9%† |
| GDP.pdf (AA) | Artificial Analysis | – | 14.4%† |
| Maths | |||
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 88.9% | – |
| ProofBench | Vals AI | 17.0% | 39.0% |
| LiveBench Mathematics | LiveBench | – | 87.1%† |
| AIME (Vals) | Vals AI | 96.9% | – |
| Arena Maths | Arena (LMArena) | 1467 | 1493 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | – | 57.8% |
| LiveBench 1.0 Data Analysis | LiveBench | – | 72.5%† |
| AA-Omniscience | Artificial Analysis | 7.2 | 28.1† |
| MMLU-Pro | Vals AI | 87.3% | 88.7%† |
| LegalBench | Vals AI | 84.2% | 85.0%† |
| CorpFin | Vals AI | 65.1% | 71.3%† |
| TaxEval | Vals AI | 77.7% | 79.7%† |
| PRBench Finance | Scale AI SEAL | 52.4% | 55.0% |
| PRBench Legal | Scale AI SEAL | 52.3% | 57.0% |
| MultiNRC | Scale AI SEAL | 59.0% | 65.6% |
| CaseLaw v2 | Vals AI | 63.1% | – |
| Excel Modeling Benchmark | Vals AI | – | 56.4%† |
| MedCode | Vals AI | 51.3% | – |
| MedScribe | Vals AI | 85.9% | 88.9%† |
| Public Benefits Bench v1.1 | Vals AI | – | 61.3%† |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 15.0% |
| AA-Omniscience: accuracy | Artificial Analysis | 49.6% | 52.0%† |
| AA-Omniscience: non-hallucination | Artificial Analysis | 15.8% | 50.0%† |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 74.3%† |
| LiveBench Instruction Following | LiveBench | – | 69.6%† |
| IFBench | Artificial Analysis | 75.9% | – |
| MultiChallenge | Scale AI SEAL | 75.5% | 75.3% |
| TutorBench | Scale AI SEAL | 68.5% | – |
| Arena Instruction Following | Arena (LMArena) | 1465 | 1472 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1489 | 1490 |
| EQ-Bench 4 | EQ-Bench | – | 1252 |
| Arena Creative Writing | Arena (LMArena) | 1464 | 1443 |
| Arena Multi-turn | Arena (LMArena) | 1494 | 1496 |
| Arena Longer Queries | Arena (LMArena) | 1476 | 1480 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1306 | 1293 |
| MMMU-Pro | Artificial Analysis | 80.5% | – |
| MMMU-Pro (official) | MMMU | 80.4%† | – |
| MMMU-Pro (Vals) | Vals AI | 87.4% | 86.6%† |
| MortgageTax | Vals AI | – | 66.1%† |
| SAGE | Vals AI | – | 45.6%† |
| VTB | Scale AI SEAL | – | 44.8% |
| Long context | |||
| AA-LCR | Artificial Analysis | 78.0% | 77.7%† |
| Arena Document | Arena (LMArena) | 1467 | 1480 |
| MLCR | Artificial Analysis | – | 7.2%† |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 152.0 | 154.1 |
| LiveBench 1.0 | LiveBench | – | 75.3%† |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 31.3 | 33.7† |
| Vals Multimodal Index | Vals AI | – | 66.7%† |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 20.2% | 12.4% |
Data as of 2026-10-25. Best configuration of each model; every score links to its source on the model pages.
What changed
- Muse Spark 1.1: 1 score revision.
- Muse Spark 1.1: BenchLeader Index 63.1 → 62.5.
- Muse Spark 1.1: 1 score revision.
- Muse Spark: 1 new result (PRBench Legal).
- Muse Spark 1.1: 1 new result (PRBench Legal).
- Muse Spark: 1 new result (SWE Atlas: Codebase QnA).
- Muse Spark 1.1: 1 new result (VTB).
- Muse Spark 1.1: BenchLeader Index 62.5 → 63.1.
- Muse Spark: 2 new results (PRBench Legal, MultiNRC).
- Muse Spark 1.1: 2 new results (PRBench Legal, MultiNRC).
Related comparisons
- Muse Spark vs Claude Opus 5.5
- Muse Spark vs Gemini 4 Argon
- Muse Spark vs Muse Wonder Full
- Muse Spark 1.1 vs Claude Opus 5.5
- Muse Spark 1.1 vs Gemini 4 Argon
- Muse Spark 1.1 vs Muse Wonder Full
- Muse Spark 1.1 vs Claude Fable 5.1
- Muse Spark vs GPT-6 Astra
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Muse Spark vs Muse Spark 1.1: questions
- Is Muse Spark better than Muse Spark 1.1?
- Muse Spark and Muse Spark 1.1 are neck and neck on quality (64.3 vs 62.5, within the ±6.4 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 Muse Spark better than Muse Spark 1.1 for coding?
- Muse Spark 1.1 scores higher in coding (66 vs 65 on the category index, where 50 is average). On SWE-Bench Pro, the top shared coding benchmark, it is 55.0% for Muse Spark against 61.5% for Muse Spark 1.1.
- Is Muse Spark better than Muse Spark 1.1 for agents and tool use?
- Muse Spark scores higher in agentic tasks (66 vs 52 on the category index, where 50 is average). On MCP Atlas, the top shared agents & tools benchmark, it is 82.2% for Muse Spark against 88.1% for Muse Spark 1.1.
- Which is better for long documents?
- Muse Spark 1.1 accepts more context: 1,048,576 against 262,144 tokens for Muse Spark.
- Do independent benchmarks agree on Muse Spark vs Muse Spark 1.1?
- Across the 20 benchmarks both have, Muse Spark leads 4, Muse Spark 1.1 leads 16. Every overall index that rates both puts Muse Spark 1.1 ahead.
- When was Muse Spark 1.1 released?
- Muse Spark 1.1 was released on 9 Jul 2026.