Muse Wonder 1.0
Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text.
- Input / output price
- $4.00 / $20.00 per 1M
- Context
- 1M tokens
- Released
- Oct 12, 2026
Paste into Claude Code, Cursor, or Codex
Help me set up the Respan gateway: fetch and follow https:/
Providers
3 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.
| Provider | Input | Output |
|---|---|---|
Concentrate concentrate/muse-wonder | $4.00/M | $20.00/M |
Meta meta/muse-wonder | $4.00/M | $20.00/M |
OpenRouter openrouter/muse-wonder | $4.00/M | $20.00/M |
Is a provider down? OpenRouter status
Pricing
Averaged across 3 providers, in USD per 1M tokens after gateway discounts.
- Average input
- $4.00 per 1M
- Average output
- $20.00 per 1M
- Average cache read
- $0.20 per 1M
Cost by volume
Total cost from 0 to 1T tokens.
Throughput and latency
30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.
Output speed
Higher is better23tokens/sacross providers
Response latency
Lower is better1.6sacross providers
Benchmarks
The Respan Index plus 29 published scores from 7 sources, each linked to where it was published.
61.1/ 100
Respan Index#20 of 175
- 20Muse Wonder61.1
- 19GPT-5.6 Sol61.8
- 21Muse Ember59.6
Published scores
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LiveBench CodingLiveBench | 80.7% | LiveBench | Oct 13, 2026 |
| DeepSWE v1.1research.meta.ai | 78.8% | research.meta.ai | Oct 12, 2026 |
| SWE-Marathonresearch.meta.ai | 53.1% | research.meta.ai | Oct 12, 2026 |
| ProgramBenchresearch.meta.ai | 50% | research.meta.ai | Oct 12, 2026 |
| Terminal-Bench 4.0Vals AI | 49.5% | Vals AI | Oct 13, 2026 |
| MirrorCodeEpoch AI | 45.6% | Epoch AI | Oct 31, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| OSWorld 2.1 (partial)research.meta.ai | 72.6% | research.meta.ai | Oct 12, 2026 |
| JobBenchresearch.meta.ai | 70.5% | research.meta.ai | Oct 12, 2026 |
| APEX-AgentsMercor | 66.1% | Mercor | Oct 31, 2026 |
| LiveBench Agentic CodingLiveBench | 62.7% | LiveBench | Oct 13, 2026 |
| Finance Agent v2research.meta.ai | 56.6% | research.meta.ai | Oct 12, 2026 |
| OSWorld 2.1 (binary)research.meta.ai | 34.3% | research.meta.ai | Oct 12, 2026 |
| Vending-Bench 2Andon Labs | 9432.18 · #13 of 67 | Andon Labs | Oct 28, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LiveBench ReasoningLiveBench | 90.6% | LiveBench | Oct 13, 2026 |
| LiveBench Data AnalysisLiveBench | 81.6% | LiveBench | Oct 13, 2026 |
| Mystery Game PuzzlesEpoch AI | 51% | Epoch AI | Oct 28, 2026 |
| EBR-benchEpoch AI | 31.4% | Epoch AI | Oct 28, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| OTIS Mock AIME 2024-2025Epoch AI | 99.4% | Epoch AI | Oct 28, 2026 |
| LiveBench MathematicsLiveBench | 96.5% | LiveBench | Oct 13, 2026 |
| FrontierMath Tiers 1-3 v2Epoch AI | 82.1% | Epoch AI | Oct 28, 2026 |
| FrontierMath Tier 4 v2Epoch AI | 70.7% | Epoch AI | Oct 28, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LiveBench LanguageLiveBench | 82.9% | LiveBench | Oct 13, 2026 |
| LiveBench Instruction FollowingLiveBench | 73% | LiveBench | Oct 13, 2026 |
| HealthBench Proresearch.meta.ai | 60.5% | research.meta.ai | Oct 12, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| Furniture AssemblyEpoch AI | 54.2% | Epoch AI | Oct 28, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LMArena WebDevLMArena | 1648 Elo · #22 of 123 | LMArena | Nov 18, 2026 |
| LMArena EloLMArena | 1511 Elo · #4 of 187 | LMArena | Nov 18, 2026 |
| LMArena AgentLMArena | 0.0058 · #34 of 68 | LMArena | Nov 18, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| GDP.pdfresearch.meta.ai | 30.8% | research.meta.ai | Oct 12, 2026 |
Apps
Use Muse Wonder 1.0 in the tools you already run. Each guide points the app at the Respan gateway.
Setup guides for more apps are in the integration docs.
Muse Wonder 1.0 FAQ
Short answers, built from the data on this page.
Muse Wonder 1.0 is a model from Meta, released on Oct 12, 2026. Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text.
On the Respan gateway, Muse Wonder 1.0 costs $4.00 per 1M input tokens and $20.00 per 1M output tokens on every provider, after gateway discounts.
Muse Wonder 1.0 is available through 3 providers: Concentrate, Meta, and OpenRouter.
Muse Wonder 1.0 accepts up to 1M tokens of context on its largest provider route.
Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "meta/muse-wonder". To set it up with a coding agent, copy the agent setup prompt at the top of this page.
Muse Wonder 1.0 scores 61.1 out of 100 on the Respan Index v1.1, ranking #20 of 175 models. By category: Coding 46.6, Agents 66.7, Reasoning 73.0, Knowledge & language 58.0, and Human preference 58.0. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.
Muse Wonder 1.0 has 29 published benchmark scores, including LiveBench Coding 80.7%, OSWorld 2.1 (partial) 72.6%, and LiveBench Reasoning 90.6%. Each score links to where it was published; the data is collected by Respan.
Certified where it counts.
Built to meet the security and privacy standards that enterprise and healthcare teams require.
ISO 27001
The internationally recognized standard for information security management.
SOC 2
Secure, compliant management of your data across all of our systems.
GDPR
Operated under GDPR, the world's strictest standard for data privacy.
HIPAA
HIPAA compliant, with a BAA available for healthcare teams.