GPT-5.4 Pro vs Muse Spark 1.1
Full profiles: GPT-5.4 Pro (OpenAI) and Muse Spark 1.1 (Meta): every benchmark, price and rank.
Updated · Data from 6 sources
As of 25 Oct 2026, GPT-5.4 Pro and Muse Spark 1.1 are neck and neck on quality (63.3 vs 62.5 on the BenchLeader Index, inside the ±6.6 margin of error), and Muse Spark 1.1 costs 34× less, which makes it the better buy for most workloads. Pick GPT-5.4 Pro for multimodal (70 vs 64); pick Muse Spark 1.1 for instruction following (72 vs 64).
GPT-5.4 Pro benchmarks 
Muse Spark 1.1 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 |
|---|---|---|---|
| Reasoning | GPT-5.4 Pro | 73 vs 68 category score | Medium |
| Knowledge | GPT-5.4 Pro | 74 vs 69 category score; MultiNRC 62.3% vs 65.6% | Medium |
| Images and other media | GPT-5.4 Pro | 70 vs 64 category score | Medium |
| Following instructions | Muse Spark 1.1 | 72 vs 64 category score; MultiChallenge 75.3% vs 69.2% | High |
| Chat quality (human preference) | – | Only Muse Spark 1.1 has an Arena rating (65 category score) | Not comparable |
| Lowest cost | Muse Spark 1.1 | $2.00 vs $67.50 per 1M blended; agentic coding session $0.092 vs $2.52 | – |
| Fastest first answer | Muse Spark 1.1 | 1.5 s (Muse Spark 1.1) vs 6.1 s (GPT-5.4 Pro) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GPT-5.4 Pro and Muse Spark 1.1 are neck and neck on quality (63.3 vs 62.5, within the ±6.6 margin).
- GPT-5.4 Pro is stronger in knowledge, multimodal, reasoning.
- Muse Spark 1.1 is stronger in instruction following, agents & tools, coding, human preference, maths.
- Muse Spark 1.1 is 34× cheaper ($2.00 vs $67.50 per 1M blended).
- Muse Spark 1.1 streams 28.6× faster (200 vs 7 tokens per second).
At a glance
| Spec | GPT-5.4 Pro | Muse Spark 1.1 |
|---|---|---|
| Developer | OpenAI | Meta |
| Released | 4 Mar 2026 | 9 Jul 2026 |
| Context window (tokens) | 1,050,000 | 1,048,576 |
| Compared at | GPT-5.4 Pro | Muse Spark 1.1 |
| Open weights | No | No |
| Price in / out / cached input per 1M | $30.00 / $180.00 / – | $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 | GPT-5.4 Pro | Muse Spark 1.1 | Leader |
|---|---|---|---|
| Reasoning | 72.5 | 67.9 | GPT-5.4 Pro |
| Coding | – | 66.3 | not comparable |
| Agents & tools | – | 51.8 | not comparable |
| Maths | – | 61.3 | not comparable |
| Knowledge | 73.6 | 69.2 | GPT-5.4 Pro |
| Instruction following | 63.8 | 71.5 | Muse Spark 1.1 |
| Human preference | – | 65.0 | not comparable |
| Multimodal | 69.8 | 63.6 | GPT-5.4 Pro |
Where sources disagree
Across the 4 benchmarks both have, GPT-5.4 Pro leads 2, Muse Spark 1.1 leads 2.
| Overall index | GPT-5.4 Pro | Muse Spark 1.1 | Leader |
|---|---|---|---|
| BenchLeader Index | 63.3 | 62.5 | neck and neck |
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 | GPT-5.4 Pro | Muse Spark 1.1 |
|---|---|---|
| Blended price per 1M tokens | $67.50 | $2.00 |
| Input / output per 1M | $30.00 / $180.00 | $1.25 / $4.25 |
| Cached input per 1M | – | $0.150 |
| Price per task: chat reply | $0.066 | $0.0018 |
| Price per task: summarize a 30-page report | $0.468 | $0.018 |
| Price per task: code edit | $0.450 | $0.014 |
| Price per task: agentic coding session | $2.52 | $0.092 |
| Price per task: structured extraction | $0.096 | $0.0034 |
| Output speed | 7 tok/s | 200 tok/s |
| Time to first answer | 6.1 s | 1.5 s |
Each model is compared at its best-ranked reasoning-effort setting. GPT-5.4 Pro scores 63.3 and takes 6 s to its first answer token; GPT-5.4 Pro (xhigh) scores 61.9 in 6 s. 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
95 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 | GPT-5.4 Pro | Muse Spark 1.1 |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 63.3 | 62.5 |
| Instruction following score | BenchLeader | 63.8 | 71.5 |
| Knowledge score | BenchLeader | 73.6 | 69.2 |
| Multimodal score | BenchLeader | 69.8 | 63.6 |
| Reasoning score | BenchLeader | 72.5 | 67.9 |
| Agents & tools score | BenchLeader | – | 51.8 |
| Coding score | BenchLeader | – | 66.3 |
| Human preference score | BenchLeader | – | 65.0 |
| Maths score | BenchLeader | – | 61.3 |
| Blended price $/M | BenchLeader | $67.50 | $2.00 |
| Output speed | BenchLeader | 7 tok/s | 200 tok/s |
| Time to first answer | BenchLeader | 6.1 s | 1.5 s |
| Context window | BenchLeader | 1,050,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 94.6%† | – |
| Humanity's Last Exam | Scale AI / CAIS | 44.3% | – |
| SimpleBench | SimpleBench | 74.1% | – |
| Arena Hard Prompts | Arena (LMArena) | – | 1510 |
| LiveBench 1.0 Reasoning | LiveBench | – | 87.7%† |
| GPQA Diamond (AA) | Artificial Analysis | – | 89.8%† |
| Humanity's Last Exam (AA) | Artificial Analysis | – | 46.2%† |
| GPQA Diamond (Vals) | Vals AI | – | 91.2%† |
| ARC-AGI-1 | ARC Prize | 94.5%† | – |
| ARC-AGI-2 | ARC Prize | 83.3%† | – |
| CritPt | Artificial Analysis | 30.0%† | 15.1%† |
| Chess Puzzles | Epoch AI Benchmarking Hub | 59.0%† | 34.0%† |
| LMCA | Epoch AI Benchmarking Hub | – | 49.9%† |
| DTBench | Epoch AI Benchmarking Hub | – | 94.4%† |
| Coding | |||
| SciCode | SciCode | – | 58.8% |
| WeirdML | WeirdML | 57.4%† | – |
| Arena Coding | Arena (LMArena) | – | 1533 |
| Arena WebDev | Arena (LMArena) | – | 1543 |
| LiveBench 1.0 Coding | LiveBench | – | 77.2%† |
| LiveCodeBench | Vals AI | – | 85.9%† |
| SWE-bench (Vals) | Vals AI | – | 82.0%† |
| SWE-Bench Pro | Scale AI SEAL | – | 61.5% |
| Code Migration | Vals AI | – | 31.1%† |
| ProgramBench | Vals AI | – | 0.0%† |
| Vibe Code Bench v1.1 | Vals AI | – | 72.2%† |
Show all 95 benchmarks
| Metric | Publisher | GPT-5.4 Pro | Muse Spark 1.1 |
|---|---|---|---|
| Coding | |||
| SWE-Bench Pro (private) | Scale AI SEAL | – | 61.5% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | – | 42.2%† |
| SWE Atlas: Test Writing | Scale AI SEAL | – | 41.5%† |
| 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 | – | 77.9%† |
| Agents & tools | |||
| APEX-Agents | Mercor | – | 31.8% |
| Arena Agent | Arena (LMArena) | – | -6.9 |
| LiveBench Agentic Coding | LiveBench | – | 58.5%† |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 69.3%† |
| MCP Atlas | Scale AI SEAL | – | 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%† |
| 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 | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 82.5%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 58.5%† | – |
| ProofBench | Vals AI | – | 39.0% |
| LiveBench Mathematics | LiveBench | – | 87.1%† |
| MathArena Apex | MathArena | 69.8%† | – |
| Arena Maths | Arena (LMArena) | – | 1493 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 46.3%† | 57.8% |
| LiveBench 1.0 Data Analysis | LiveBench | – | 72.5%† |
| AA-Omniscience | Artificial Analysis | – | 28.1† |
| MMLU-Pro | Vals AI | – | 88.7%† |
| LegalBench | Vals AI | – | 85.0%† |
| CorpFin | Vals AI | – | 71.3%† |
| TaxEval | Vals AI | – | 79.7%† |
| PRBench Finance | Scale AI SEAL | – | 55.0% |
| PRBench Legal | Scale AI SEAL | – | 57.0% |
| MultiNRC | Scale AI SEAL | 62.3% | 65.6% |
| Excel Modeling Benchmark | Vals AI | – | 56.4%† |
| MedScribe | Vals AI | – | 88.9%† |
| Public Benefits Bench v1.1 | Vals AI | – | 61.3%† |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 15.0% |
| AA-Omniscience: accuracy | Artificial Analysis | – | 52.0%† |
| AA-Omniscience: non-hallucination | Artificial Analysis | – | 50.0%† |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 74.3%† |
| LiveBench Instruction Following | LiveBench | – | 69.6%† |
| MultiChallenge | Scale AI SEAL | 69.2% | 75.3% |
| TutorBench | Scale AI SEAL | 56.6% | – |
| Arena Instruction Following | Arena (LMArena) | – | 1472 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1490 |
| EQ-Bench 4 | EQ-Bench | – | 1252 |
| Arena Creative Writing | Arena (LMArena) | – | 1443 |
| Arena Multi-turn | Arena (LMArena) | – | 1496 |
| Arena Longer Queries | Arena (LMArena) | – | 1480 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | – | 1293 |
| VISTA | Scale AI SEAL | 53.9% | – |
| MMMU-Pro (Vals) | Vals AI | – | 86.6%† |
| MortgageTax | Vals AI | – | 66.1%† |
| SAGE | Vals AI | – | 45.6%† |
| VTB | Scale AI SEAL | – | 44.8% |
| Long context | |||
| AA-LCR | Artificial Analysis | – | 77.7%† |
| Arena Document | Arena (LMArena) | – | 1480 |
| MLCR | Artificial Analysis | – | 7.2%† |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 159.6 | 154.1 |
| LiveBench 1.0 | LiveBench | – | 75.3%† |
| AA Intelligence Index v4.3.2 | Artificial Analysis | – | 33.7† |
| Vals Multimodal Index | Vals AI | – | 66.7%† |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 14.8% | 12.4% |
| MASK | Scale AI SEAL | 91.7% | – |
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.
- GPT-5.4 Pro: 1 score revision.
- Muse Spark 1.1: 1 score revision.
- Muse Spark 1.1: 1 new result (PRBench Legal).
- Muse Spark 1.1: 1 new result (VTB).
- GPT-5.4 Pro: BenchLeader Index 62.1 → 63.6.
- Muse Spark 1.1: BenchLeader Index 62.5 → 63.1.
- GPT-5.4 Pro: 1 new result (MultiNRC).
- Muse Spark 1.1: 2 new results (PRBench Legal, MultiNRC).
Related comparisons
- GPT-5.4 Pro vs Claude Opus 5.5
- GPT-5.4 Pro vs Gemini 4 Argon
- GPT-5.4 Pro 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
- GPT-5.3-Codex vs Muse Spark 1.1
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
GPT-5.4 Pro vs Muse Spark 1.1: questions
- Is GPT-5.4 Pro better than Muse Spark 1.1?
- GPT-5.4 Pro and Muse Spark 1.1 are neck and neck on quality (63.3 vs 62.5, within the ±6.6 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.
- Which is cheaper, GPT-5.4 Pro or Muse Spark 1.1, and will that change?
- Muse Spark 1.1 is cheaper today: $2.00 against $67.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Which is better for long documents?
- Both take about 1.1M tokens: GPT-5.4 Pro 1,050,000, Muse Spark 1.1 1,048,576.
- Which has lower latency?
- Muse Spark 1.1 answers first: 1.5 s against 6.1 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Which is faster, GPT-5.4 Pro or Muse Spark 1.1?
- Muse Spark 1.1 streams faster: 200 against 7 output tokens per second.
- Do independent benchmarks agree on GPT-5.4 Pro vs Muse Spark 1.1?
- Across the 4 benchmarks both have, GPT-5.4 Pro leads 2, Muse Spark 1.1 leads 2.
- When was Muse Spark 1.1 released?
- Muse Spark 1.1 was released on 9 Jul 2026.