Models / Muse Wonder
Meta
Muse Wonder
Muse Wonder is best for user surveys and product listings; pick it when quality matters more than cost.
Results 13 Oct 2026 · 84 results · 5 sources
- Price per million tokens
- $4.00 in · $20.00 out
Meta's list price, 13 Oct 2026 · compare prices$0.20 cached input - Speed
- Not measured
- Intelligence Index
- Not measured
- Spring Prompt overall
- 4th of 23 score 70 of 100 · how it works
- Developer
- Meta, based in the US
Weights not published - Released
- 12 Oct 2026
1,048,576 tokens of context - Pricing and availability
- Standard Wonder. Muse Wonder Full is offered separately through Muse Code subscriptions.
Where it stands among the models you could choose
Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.
| Among | Intelligence Index | Spring Prompt overall | Price, blended | Speed |
|---|---|---|---|---|
| Models available today | – | 4th of 23Claude Opus 5.5 | $8.00 blended, 92nd percentilecheapest: Mistral Nemo | – |
| Models priced $3–10 per million tokens | – | 3rd of 11Claude Opus 5.5 | sets this group | – |
| Models from developers based in the US | – | 4th of 15Claude Opus 5.5 | $8.00 blended, 84th percentilecheapest: Llama 3.1 8B Instruct | – |
The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.
Measured cost per task
API cost recorded in our benchmark runs, excluding judging.
| Task | Benchmark | Cost per task |
|---|---|---|
| Product listing | CatalogBench | ≈ $0.036 |
| Deck | DeckBench | ≈ $0.182 |
| Survey plan and analysis | SurveyBench | ≈ $0.354 |
| Game of blitz chess | BulletBench | ≈ $0.267 |
How good is it, and for what?
Each use case ranks the available models its benchmarks measured, then those priced $3–10 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.
Product listingsTurning a sparse product feed and photos into listings that can go live
| Benchmark | Muse Wonder | Rank among available | Best available |
|---|---|---|---|
| CatalogBenchCatalogBench: reliably publish-ready | 66.1% | 3rd of 21 | GPT-6 Astra 75.0% |
| CatalogBenchCatalogBench (sales brief): reliably publish-ready | 51.8% | 4th of 21 | GPT-6 Astra 69.6% |
Decks from an analysisTurning a finished analysis into a deck you could present as it is
| Benchmark | Muse Wonder | Rank among available | Best available |
|---|---|---|---|
| DeckBenchDeckBench: deck rating | 1,273 | 8th of 20 | GPT-6 Astra 1,492 |
User surveysPlanning a user survey and reading its results without being misled
| Benchmark | Muse Wonder | Rank among available | Best available |
|---|---|---|---|
| SurveyBenchSurveyBench: analysis rating | 1,460 | 2nd of 9 | Claude Opus 5.5 1,499 |
Marketing planningPlanning a year of ad spend without overspending
| Benchmark | Muse Wonder | Rank among available | Best available |
|---|---|---|---|
| ROASBenchROASBench: overall score | 38.9 | 10th of 20 | GPT-6 Astra 56.6 |
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
- tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
- Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
- OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
- Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
- Artificial Analysis: pending. Its latest results are dated 8 Oct 2026, before this model was released.
- Vending-Bench 2: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
- APEX-Agents: pending. Its latest results are dated 1 Oct 2026, before this model was released.
- GDP.pdf: pending. Its latest results are dated 1 Oct 2026, before this model was released.
- Remote Labor Index: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Reasoning and knowledgeHard questions across science, maths and general knowledge
- Artificial Analysis: pending. Its latest results are dated 8 Oct 2026, before this model was released.
WritingWhat people prefer in blind comparisons, and judged writing quality
- Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
- UGI Leaderboard: pending. Its latest results are dated 2 Oct 2026, before this model was released.
Sticking to the factsSummarising without inventing things, and factual answers
- Vectara Hallucination Leaderboard: pending. Its latest results are dated 22 Sep 2026, before this model was released.
- Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
- SimpleQA Verified (Epoch AI): pending. Its latest results are dated 29 Sep 2026, before this model was released.
Speed under pressureGood decisions against a real clock (fast chess)
| Benchmark | Muse Wonder | Rank among available | Best available |
|---|---|---|---|
| BulletBenchBulletBench Bullet 60s: ladder Elo | 409 | 13th of 22 | Gemini 3.8 Flash 1,068 |
Nearest alternatives
- The most capable open-weights modelMiMo-V2.6-Pro$0.54 blended · Intelligence Index 46.3
Against Muse Wonder at $8.00 blended. "Similar intelligence" means within 4 points or better.
Against the alternatives
The models you would most likely weigh it against: the leaders of the groups above.
Scroll sideways to see every alternative.
| Muse Wonder | Claude Opus 5.5 highest-scoring overall at $3–10 per million tokens | GPT-6 Astra just above overall | GPT-6 Sol just below overall | Muse Spark 1.3 Meta's best other model | |
|---|---|---|---|---|---|
| Price per million tokens | $8.00 | $8.00 | $20.00 | $4.00 | $2.00 |
| Tokens a second | – | 97 | 52 | – | 114 |
| Intelligence Index | – | 57.6 | 52.7 | 47.6 | 48.1 |
| Spring Prompt overall | 70 | 85 | 78 | 69 | 64 |
| Rank among available models, by use case | |||||
| Product listings | 4th of 21 | 7th | 1st | 3rd | 10th |
| Decks from an analysis | 8th of 20 | 4th | 1st | 2nd | 13th |
| User surveys | 2nd of 9 | 1st | – | – | – |
| Marketing planning | 10th of 20 | 5th | 1st | 2nd | 13th |
| Agents and tool use | pending | 2nd | 4th | 5th | 9th |
| Professional work | pending | 3rd | 5th | 13th | 8th |
| Reasoning and knowledge | pending | 1st | 3rd | 11th | 8th |
| Writing | pending | 4th | 19th | 51st | 5th |
| Sticking to the facts | pending | 2nd | 24th | 49th | 11th |
| Speed under pressure | 12th of 24 | – | 16th | 10th | 21st |
| Side by side → | Side by side → | Side by side → | Side by side → | ||
Shaded figures are better than Muse Wonder's. A dash means no result.
Research and examples
- Claude Opus 5.5 vs Muse Wonder Full: Eight Crafts, and Opus Still Wins Most
- Muse Wonder review: good analysis, work that still needs checking
In the playground: our pick in 0 of 8 tests, 64% of checks passed →
- Paint a lemon on a blue plate as a thick oil painting on canvas (3 of 4 checks)
- Bend a neon sign that says "open late" from glass tubing (3 of 4 checks)
- Embroider a bee on a linen patch (3 of 4 checks)
Where it has been measured
3rd highest-scoring overall of the 11 models priced $3–10 per million tokens. Among models available today, in the top quarter (by percentile) for user surveys and product listings. Results for agents and tool use, professional work, reasoning and knowledge, writing, and sticking to the facts are pending: those sources last updated before this model was released.
Not timed: Artificial Analysis, our source for speed, has not tested it
- BulletBench10 results
- CatalogBench30 results
- DeckBench16 results
- ROASBench9 results
- SurveyBench19 results
- APEX-AgentsPending: latest results 1 Oct 2026
- Arena (formerly LMArena)Pending: latest results 2 Oct 2026
- Artificial AnalysisPending: latest results 8 Oct 2026
- Berkeley Function Calling Leaderboard (BFCL) V4Pending: latest results 16 Dec 2025
- GDP.pdfPending: latest results 1 Oct 2026
- Microsoft STATE-BenchPending: latest results 1 Jul 2026
- OpenHands IndexPending: latest results 30 Jun 2026
- Remote Labor IndexPending: latest results 1 Oct 2026
- SimpleQA Verified (Epoch AI)Pending: latest results 29 Sep 2026
- tau2-benchPending: latest results 3 Aug 2026
- UGI LeaderboardPending: latest results 2 Oct 2026
- Vectara Hallucination LeaderboardPending: latest results 22 Sep 2026
- Vending-Bench 2Pending: latest results 1 Oct 2026
Every result
84 published results from 5 sources, each in the source's own units, with the configuration that produced it.
BulletBench · measured by Spring Prompt · 10 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| BulletBench Blitz 3+2: cost per gameUS dollars, lower is better | $0.27 | 12 of 15 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| BulletBench Blitz 3+2: games lost on time% of games, lower is better | 75.0% | 15 of 15 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| BulletBench Blitz 3+2: invalid moves% of moves, lower is better | 0.0% | 1 of 15 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| BulletBench Blitz 3+2: ladder Eloladder Elo, higher is better | 358 0–600 |
14 of 15 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| BulletBench Blitz 3+2: median move timemilliseconds, lower is better | 7.1 s | 13 of 15 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| BulletBench Bullet 60s: cost per gameUS dollars, lower is better | $0.0663 | 32 of 36 | Muse Wonder (minimal reasoning) minimal reasoning |
12 Oct 2026 |
| BulletBench Bullet 60s: games lost on time% of games, lower is better | 58.3% | 25 of 36 | Muse Wonder (minimal reasoning) minimal reasoning |
12 Oct 2026 |
| BulletBench Bullet 60s: invalid moves% of moves, lower is better | 0.0% | 1 of 36 | Muse Wonder (minimal reasoning) minimal reasoning |
12 Oct 2026 |
| BulletBench Bullet 60s: ladder Eloladder Elo, higher is better | 409 100–609 |
21 of 36 | Muse Wonder (minimal reasoning) minimal reasoning |
12 Oct 2026 |
| BulletBench Bullet 60s: median move timemilliseconds, lower is better | 1.9 s | 27 of 36 | Muse Wonder (minimal reasoning) minimal reasoning |
12 Oct 2026 |
Not shown: Not your cost: prices are those charged on the run date.
CatalogBench · measured by Spring Prompt · 30 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| CatalogBench (sales brief): channel rules broken% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): claims to check% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): content quality% of checks, higher is better | 98.5% | 3 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): failed outputs% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): missing UK information% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): not findable% of products, lower is better | 3.6% | 8 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): publish-ready listings% of products, higher is better | 67.3% | 5 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): reliably publish-ready% of products, higher is better | 51.8% | 4 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): unsupported claims% of products, lower is better | 17.9% | 5 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): unsupported claimsclaims per product, lower is better | 0.36 | 5 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): wrong attributes% of products, lower is better | 14.3% | 4 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench (sales brief): wrong category or variant% of products, lower is better | 2.4% | 10 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: channel compliance% of products, higher is better | 99.4% | 14 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: channel rules broken% of products, lower is better | 0.6% | 11 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: claims to check% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: conflicts caught% of conflicts, higher is better | 100.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: content quality% of checks, higher is better | 98.3% | 3 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: cost per productUS dollars, lower is better | $0.0361 | 18 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: decision accuracy% of decisions, higher is better | 98.3% | 3 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: failed outputs% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: field accuracy% of missing fields, higher is better | 96.1% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: invented values% of filled values, lower is better | 1.4% | 11 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: missing UK information% of products, lower is better | 0.0% | 1 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: not findable% of products, lower is better | 3.6% | 4 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: publish-ready listings% of products, higher is better | 75.0% | 3 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: reliably publish-ready% of products, higher is better | 66.1% | 3 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: unsupported claims% of products, lower is better | 7.1% | 4 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: unsupported claimsclaims per product, lower is better | 0.21 | 4 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: wrong attributes% of products, lower is better | 13.1% | 2 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| CatalogBench: wrong category or variant% of products, lower is better | 2.4% | 10 of 22 | Muse Wonder provider default reasoning |
12 Oct 2026 |
Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
DeckBench · measured by Spring Prompt · 16 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| DeckBench: accurate decks% of tasks, higher is better | 100.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: caveat dropped% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: clean layout% of tasks, higher is better | 16.7% | 18 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: cost per deckUS dollars, lower is better | $0.18 | 14 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: deck ratingrating, higher is better | 1,273 1,175–1,385 |
8 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: design quality% of the maximum, higher is better | 67.1% | 9 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: draft figure quoted% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: findings missing% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: head-to-head win rate% of comparisons, higher is better | 63.2% | 10 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: layout defects% of tasks, lower is better | 83.3% | 18 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: misleading metric used% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: presentable decks% of tasks, higher is better | 16.7% | 10 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: recommendation late or wrong% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: slides needing work% of tasks, lower is better | 66.7% | 15 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: unsupported claims% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| DeckBench: unsupported numbers% of tasks, lower is better | 0.0% | 1 of 20 | Muse Wonder provider default reasoning |
12 Oct 2026 |
Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.
ROASBench · measured by Spring Prompt · 9 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| ROASBench: audience scorescore out of 100, higher is better | 54.7 | 12 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: business scorescore out of 100, higher is better | 34.5 | 10 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: consistency scorescore out of 100, higher is better | 25.9 | 12 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: contribution profitsimulated US dollars, higher is better | $489,176 | 11 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: cost of a runUS dollars, lower is better | $0.57 | 16 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: months over budgetmonths of 12, lower is better | 0 | 1 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: overall scorescore out of 100, higher is better | 38.9 | 10 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: planning scorescore out of 100, higher is better | 55.0 | 4 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| ROASBench: return on ad spendprofit per $1 spent, higher is better | $0.41 | 11 of 21 | Muse Wonder provider default reasoning |
12 Oct 2026 |
Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.
SurveyBench · measured by Spring Prompt · 19 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| SurveyBench: analysis ratingrating, higher is better | 1,460 1,400–1,545 |
2 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: analysis scorepoints of 100, higher is better | 96.2 | 2 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: cost per taskUS dollars, lower is better | $0.35 | 8 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: head-to-head win rate% of comparisons, higher is better | 67.9% | 3 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: leading or double-barrelled questions% of tasks, lower is better | 44.4% | 5 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: mistaken next steps% of tasks, lower is better | 0.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: numbers right% of answers, higher is better | 100.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: objectives unanswerable% of tasks, lower is better | 11.1% | 5 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: precision ignored% of tasks, lower is better | 22.2% | 4 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: research scorepoints of 100, higher is better | 93.6 | 3 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: sound research% of tasks, higher is better | 33.3% | 3 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: stakeholder's view presumed% of tasks, lower is better | 11.1% | 4 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: survey plan scorepoints of 100, higher is better | 87.8 | 5 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: survey structure% of tasks, lower is better | 33.3% | 8 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: traps handled% of traps, higher is better | 100.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: traps missed% of tasks, lower is better | 0.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: unsupported findings% of tasks, lower is better | 11.1% | 2 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: wrong numbers% of tasks, lower is better | 0.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
| SurveyBench: wrong recommendation% of tasks, lower is better | 0.0% | 1 of 9 | Muse Wonder provider default reasoning |
12 Oct 2026 |
Not shown: Not your customers or your data: invented organisations with simulated respondents, nine public tasks, one run per model per task at the provider's default reasoning setting.