whatmodel.sh / Muse Wonder
Muse Wonder
↗ share
+ contribute
Muse Wonder is a closed-weights language model from Meta, released October 2026.
$ how it compares
The LLMs ranked either side of Muse Wonder, out of 830 with a score, then more from Meta.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 14 |  GPT-5.4 Pro | 81±1 | api $67.50/M | ||
| 15 | GPT-5.5 | 80±1 | api $11.25/M | ||
| 16 | Muse Wonder | 79±2 | api $8.00/M | ||
| 17 |  MiMo-V2.6-Pro | 79±1 | api $0.54/M +local | ||
| 18 |  Hy4 preview (770B-A49B) | 79±1 | api $1.25/M +local | ||
| more from Meta | |||||
| 5 | Muse Wonder Full | 84±3 | Muse Code only | ||
| 25 | Muse Spark 1.3 | 77±1 | api $2.00/M | ||
| 44 | Muse Spark 1.2 | 75±1 | api $2.00/M | ||
$ what it's good at
agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #31 of 302 * · 16% of its score · 2 tests
Trails GLM-5.3-Flash, Grok 4.7, Qwen3.8 Max (0902) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JobBench lab | 70.5% ✎ | █·█····█···█ beats 2/3 | 10% |
| OSWorld 2.1 (partial score) lab | 72.6% ✎ | ██·····█··██ beats 0/4 | 6.2% |
other tests3 not in a task · 83% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| EBR-bench (Epoch) | 31.4% ✎ | ▄▂·▆█▂▆▆▆▆·▄ beats 15/24 | 72% |
| ProgramBench (Almost@1) lab | 50.0% ✎ | ██·······█·█ beats 2/3 | 5.7% |
| HealthBench Professional lab | 60.5% ✎ | ▂·▂··▂▂▂█▄▂▄ beats 6/13 | 4.7% |
$ where to run it
Through an API
api · $8.00/M · $4 in / $20 out per 1M tokens · 1M context · dev.meta.ai ↗ · price sourceUSD per 1M tokens, Meta Developer Platform Standard mode for muse-wonder-1.0; cached input $0.20/1M excluded; Contributor mode is not available for Muse Wonder; web-search grounding billed separately ($2.50 per 1,000 searches); page states no long-context premium. Currency: USD. Meta official pricing page, observed 2026-10-14.
$ how the score was worked out · Epoch AI — internal evaluatio… 72%, Meta vendor-reported benchmar… 26%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Epoch AI — internal evaluations | 1 | 72% |
 Meta vendor-reported benchmark scores for its own Muse models (Muse Wonder model card, Muse Spark 1.3 evaluation methodology) · lab-reported | 4 | 26% |
| prior · pull toward the middle | 1% | |
| weighted average | = 79.3 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json