GPT-6.1 Terra vs Muse Wonder
Full profiles: GPT-6.1 Terra (OpenAI) and Muse Wonder (Meta): every benchmark, price and rank.
Updated · Data from 7 sources
As of 27 Oct 2026, GPT-6.1 Terra (max) and Muse Wonder (max) are neck and neck on quality (65.9 vs 67.3 on the BenchLeader Index, inside the ±8.2 margin of error), and GPT-6.1 Terra (max) costs 4.0× less, which makes it the better buy for most workloads. GPT-6.1 Terra leads in every category we measure both in.
GPT-6.1 Terra benchmarks 
Muse Wonder 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-6.1 Terra | 86 vs 81 category score; CritPt 28.0% vs 28.3% | Medium |
| Knowledge | GPT-6.1 Terra | 75 vs 73 category score; AA-Omniscience 29.8 vs 25.5 | Medium |
| Long documents | GPT-6.1 Terra | 66 vs 65 category score; AA-LCR 82.7% vs 80.7% | Medium |
| Chat quality (human preference) | – | Only Muse Wonder has an Arena rating (71 category score) | Not comparable |
| Lowest cost | GPT-6.1 Terra | $2.00 vs $8.00 per 1M blended; agentic coding session $0.080 vs $0.320 | – |
| Fastest first answer | Muse Wonder | 41.2 s (Muse Wonder (max)) vs 131.8 s (GPT-6.1 Terra (max)) 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-6.1 Terra (max) and Muse Wonder (max) are neck and neck on quality (65.9 vs 67.3, within the ±8.2 margin).
- GPT-6.1 Terra (max) is stronger in knowledge, long context, multimodal, reasoning.
- Muse Wonder (max) is stronger in composite, agents & tools, coding, human preference, maths.
- GPT-6.1 Terra (max) is 4.0× cheaper ($2.00 vs $8.00 per 1M blended).
- Muse Wonder (max) streams 2.2× faster (202 vs 91 tokens per second).
At a glance
| Spec | GPT-6.1 Terra | Muse Wonder |
|---|---|---|
| Developer | OpenAI | Meta |
| Released | 26 Oct 2026 | 12 Oct 2026 |
| Context window (tokens) | 1,050,000 | 1,048,576 |
| Compared at | GPT-6.1 Terra (max) | Muse Wonder (max) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $1.00 / $5.00 / $0.050 | $4.00 / $20.00 / $0.200 |
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-6.1 Terra (max) | Muse Wonder (max) | Leader |
|---|---|---|---|
| Reasoning | 86.2 | 80.8 | GPT-6.1 Terra |
| Coding | – | 66.0 | not comparable |
| Agents & tools | – | 55.8 | not comparable |
| Maths | – | 67.8 | not comparable |
| Knowledge | 74.5 | 72.6 | GPT-6.1 Terra |
| Human preference | – | 70.7 | not comparable |
| Multimodal | 67.5 | – | not comparable |
| Long context | 66.3 | 65.2 | GPT-6.1 Terra |
| Composite | 82.8 | 87.5 | Muse Wonder |
Where sources disagree
Across the 13 benchmarks both have, GPT-6.1 Terra leads 5, Muse Wonder leads 8. Every overall index that rates both puts Muse Wonder ahead.
| Overall index | GPT-6.1 Terra (max) | Muse Wonder (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.9 | 67.3 | neck and neck |
| AA Intelligence Index v4.3.2 | 48.1 | 52.4 | Muse Wonder |
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-6.1 Terra (max) | Muse Wonder (max) |
|---|---|---|
| Blended price per 1M tokens | $2.00 | $8.00 |
| Input / output per 1M | $1.00 / $5.00 | $4.00 / $20.00 |
| Cached input per 1M | $0.050 | $0.200 |
| Price per task: chat reply | $0.0019 | $0.0076 |
| Price per task: summarize a 30-page report | $0.015 | $0.060 |
| Price per task: code edit | $0.013 | $0.054 |
| Price per task: agentic coding session | $0.080 | $0.320 |
| Price per task: structured extraction | $0.0030 | $0.012 |
| Cost to run a full benchmark suite | $0.403 | $3.27 |
| Output speed | 91 tok/s | 202 tok/s |
| Time to first answer | 131.8 s | 41.2 s |
| Time to a complete answer | 137.3 s | 43.6 s |
Per token is not per task: on Artificial Analysis' full suite GPT-6.1 Terra (max) cost $0.403 and Muse Wonder (max) $3.27, because Muse Wonder (max) used about 2.0× as many tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. GPT-6.1 Terra (max) scores 65.9 and takes 132 s to its first answer token; GPT-6.1 Terra (xhigh) scores 65.0 in 18 s, GPT-6.1 Terra (high) scores 64.7 in 7 s. Muse Wonder (max) scores 67.3 and takes 41 s to its first answer token; Muse Wonder scores 64.0 in 3 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
87 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-6.1 Terra (max) | Muse Wonder (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.9 | 67.3 |
| Composite score | BenchLeader | 82.8 | 87.5 |
| Knowledge score | BenchLeader | 74.5 | 72.6 |
| Long context score | BenchLeader | 66.3 | 65.2 |
| Multimodal score | BenchLeader | 67.5 | – |
| Reasoning score | BenchLeader | 86.2 | 80.8 |
| Agents & tools score | BenchLeader | – | 55.8 |
| Coding score | BenchLeader | – | 66.0 |
| Human preference score | BenchLeader | – | 70.7 |
| Maths score | BenchLeader | – | 67.8 |
| Blended price $/M | BenchLeader | $2.00 | $8.00 |
| Output speed | BenchLeader | 91 tok/s | 202 tok/s |
| Time to first answer | BenchLeader | 131.8 s | 41.2 s |
| Context window | BenchLeader | 1,050,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| Arena Hard Prompts | Arena (LMArena) | – | 1522 |
| LiveBench 2.0 Reasoning | LiveBench | 88.7%† | 90.6%† |
| LiveBench 1.0 Reasoning | LiveBench | – | 90.6% |
| Humanity's Last Exam (AA) | Artificial Analysis | 46.3% | 50.1% |
| CritPt | Artificial Analysis | 28.0% | 28.3% |
| MysteryMechanism | Vals AI | – | 27.5% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | – | 51.0% |
| LMCA | Epoch AI Benchmarking Hub | – | 56.8%† |
| DTBench | Epoch AI Benchmarking Hub | – | 95.2%† |
| Coding | |||
| SciCode | SciCode | – | 60.3% |
| Arena Coding | Arena (LMArena) | – | 1545 |
| Arena WebDev | Arena (LMArena) | – | 1651 |
| LiveBench 2.0 Coding | LiveBench | 90.8%† | 86.2%† |
| LiveBench 1.0 Coding | LiveBench | – | 80.7% |
| SciCode (AA) | Artificial Analysis | 55.1% | 60.3% |
| IOI | Vals AI | – | 66.9% |
| Code Migration | Vals AI | – | 62.5% |
| Code Migration v1.1 | Vals AI | – | 72.4% |
| SWE-bench Multimodal | Vals AI | – | 45.6% |
| Vibe Code Bench 1-100 | Vals AI | – | 26.6% |
| Vibe Code Bench v1.1 | Vals AI | – | 83.1% |
| CursorBench | Cursor | – | 50.4% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 55.6% | 52.0% |
| FrontierCode 1.1 (AA) | Artificial Analysis | 36.0% | 46.5% |
| Agents & tools | |||
| Arena Agent | Arena (LMArena) | – | 3.3 |
Show all 87 benchmarks
| Metric | Publisher | GPT-6.1 Terra (max) | Muse Wonder (max) |
|---|---|---|---|
| Agents & tools | |||
| LiveBench Long Agentic | LiveBench | 48.6%† | 52.6%† |
| LiveBench Browser | LiveBench | 93.1%† | 60.0%† |
| LiveBench Agentic Coding | LiveBench | – | 62.7% |
| HiL-Bench | Scale AI SEAL | – | 79.7%† |
| GDPval-AA v2.1 | Artificial Analysis | – | 63.5% |
| CUA-bench | Vals AI | – | 7.2% |
| CyberBench | Vals AI | – | 80.4% |
| Finance Agent v2 | Vals AI | – | 56.6% |
| Finance Agent v2.1 | Vals AI | – | 65.5% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 23.8% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 17.1% |
| Legal Research Bench | Vals AI | – | 50.5% |
| Tax Agent Bench | Vals AI | – | 36.3% |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 49.5% |
| Terminal-Bench Science | Vals AI | – | 22.9% |
| Vending-Bench 2 | Andon Labs | – | 9432.2† |
| AutomationBench | Artificial Analysis | 60.2% | 68.1% |
| GDP.pdf (AA) | Artificial Analysis | 28.6% | 30.6% |
| Harvey LAB-AA v1.1 | Harvey | – | 10.0% |
| AA-Briefcase v1.1 | Artificial Analysis | 1402 | 1659 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 57.6% | 25.7% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 82.1% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 70.7% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | – | 99.4% |
| ProofBench | Vals AI | – | 84.0%† |
| LiveBench Mathematics | LiveBench | – | 96.5% |
| Arena Maths | Arena (LMArena) | – | 1496 |
| Knowledge | |||
| LiveBench 2.0 Data Analysis | LiveBench | 82.3%† | 81.6%† |
| LiveBench Research | LiveBench | 73.9%† | 88.0%† |
| LiveBench 1.0 Data Analysis | LiveBench | – | 81.6% |
| AA-Omniscience | Artificial Analysis | 29.8 | 25.5 |
| PRBench Finance | Scale AI SEAL | – | 58.7%† |
| PRBench Legal | Scale AI SEAL | – | 58.4%† |
| BioMysteryBench | Vals AI | – | 67.0% |
| Excel Modeling Benchmark | Vals AI | – | 67.1% |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 72.5% |
| Public Benefits Bench v1.1 | Vals AI | – | 70.2% |
| EBR-bench | Epoch AI Benchmarking Hub | – | 31.4% |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 30.8% |
| AA-Omniscience: accuracy | Artificial Analysis | 49.1% | 41.1% |
| AA-Omniscience: non-hallucination | Artificial Analysis | 62.0% | 73.5% |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 82.9% |
| LiveBench Instruction Following | LiveBench | – | 73.0% |
| Arena Instruction Following | Arena (LMArena) | – | 1498 |
| Human preference | |||
| Arena Text | Arena (LMArena) | – | 1509 |
| EQ-Bench 4 | EQ-Bench | – | 1324† |
| Arena Creative Writing | Arena (LMArena) | – | 1492 |
| Arena Multi-turn | Arena (LMArena) | – | 1519 |
| Arena Longer Queries | Arena (LMArena) | – | 1502 |
| Multimodal | |||
| LiveBench Design | LiveBench | 82.3%† | 75.1%† |
| MMMU-Pro | Artificial Analysis | 84.7% | – |
| Long context | |||
| AA-LCR | Artificial Analysis | 82.7% | 80.7% |
| MLCR | Artificial Analysis | – | 22.2% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | – | 161.8† |
| LiveBench 2.0 | LiveBench | 77.2%† | 74.2%† |
| LiveBench 1.0 | LiveBench | – | 81.2% |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 48.1 | 52.4 |
| Vals Index | Vals AI | – | 64.9 |
| Vals Finance Index | Vals AI | – | 61.2% |
| Vals Science Index | Vals AI | – | 39.1% |
| Vals Router Index | Vals AI | – | 64.9%† |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 18.2%† | – |
Data as of 2026-10-27. Best configuration of each model; every score links to its source on the model pages.
What changed
- Muse Wonder (max): cost per task moved from $8.63/M to $3.27/M (62% down).
- Muse Wonder (max): cost per task moved from $4.54/M to $8.63/M (90% up).
- Muse Wonder (max): 8 new results (MLCR, Code Migration v1.1, Excel Modeling Benchmark v1.1…).
- Muse Wonder (max): 2 new results (FrontierCode 1.1 (AA), Terminal-Bench-Science 0.1 (AA)).
- Muse Wonder (max): 1 new result (Arena Maths); 2 score revisions.
- Muse Wonder (max): 3 score revisions.
- Muse Wonder (max): BenchLeader Index 67.1 → 67.6.
- Muse Wonder (max): 1 new result (CursorBench).
- Muse Wonder (max): 1 new result (Mystery Game Puzzles); 5 score revisions.
- Muse Wonder (max): BenchLeader Index 67.6 → 67.1.
Related comparisons
- GPT-6.1 Terra vs Claude Opus 5.5
- GPT-6.1 Terra vs Claude Fable 5.1
- GPT-6.1 Terra vs GPT-6 Astra
- Muse Wonder vs Claude Opus 5.5
- Muse Wonder vs Claude Fable 5.1
- Muse Wonder vs GPT-6 Astra
- Muse Wonder vs Muse Wonder Full
- GPT-6.1 Sol vs Muse Wonder
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
GPT-6.1 Terra vs Muse Wonder: questions
- Is GPT-6.1 Terra better than Muse Wonder?
- GPT-6.1 Terra (max) and Muse Wonder (max) are neck and neck on quality (65.9 vs 67.3, within the ±8.2 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-27, so check the category scores for your use.
- Which is cheaper, GPT-6.1 Terra or Muse Wonder, and will that change?
- GPT-6.1 Terra is cheaper today: $2.00 against $8.00 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?
- GPT-6.1 Terra scores higher on long-context benchmarks (66 vs 65). Both take about 1.1M tokens: GPT-6.1 Terra 1,050,000, Muse Wonder 1,048,576.
- Which has lower latency?
- Muse Wonder (max) answers first: 41.2 s against 131.8 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-6.1 Terra or Muse Wonder?
- Muse Wonder streams faster: 202 against 91 output tokens per second.
- Do independent benchmarks agree on GPT-6.1 Terra vs Muse Wonder?
- Across the 13 benchmarks both have, GPT-6.1 Terra leads 5, Muse Wonder leads 8. Every overall index that rates both puts Muse Wonder ahead.
- When was GPT-6.1 Terra released?
- GPT-6.1 Terra was released on 26 Oct 2026.