whatmodel.sh / Muse Wonder Full
Muse Wonder Full
↗ share
+ contribute
Muse Wonder Full is a closed-weights language model from Meta, released October 2026.
Catalogue identity notes
The most capable Muse Wonder model. Meta offers it to Muse Code subscribers only, not on the Meta Model API (announcement, 2026-10-12).
$ how it compares
The LLMs ranked either side of Muse Wonder Full, out of 830 with a score, then more from Meta.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 3 |  GPT-6.1 Sol | 85±1 | api $4.00/M | ||
| 4 |  Claude Fable 5.1 | 84±1 | api $20.00/M | ||
| 5 | Muse Wonder Full | 84±3 | Muse Code only | ||
| 6 | Claude Sonnet 5.5 | 84±1 | api $4.00/M | ||
| 7 | Claude Opus 5 | 83±1 | api $10.00/M | ||
| more from Meta | |||||
| 16 | Muse Wonder | 79±2 | api $8.00/M | ||
| 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▁▁▃▆█▆▅▅▅▄▃▁ #17 of 302 * · 55% of its score · 2 tests
Trails GLM-5.3, Claude Haiku 5.5, DeepSeek V4.1 Flash here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| JobBench lab | 75.3% ✎ | █·█····█···█ beats 3/3 | 33% |
| OSWorld 2.1 (partial score) lab | 78.4% ✎ | ██·····█··██ beats 2/4 | 22% |
other tests2 not in a task · 40% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| ProgramBench (Almost@1) lab | 59.0% ✎ | ██·······█·█ beats 3/3 | 22% |
| HealthBench Professional lab | 62.3% ✎ | ▂·▂··▂▂▂█▄▂▄ beats 10/13 | 18% |
$ where to run it
Only in Muse Code, Meta's own app: there's no API or download. research.meta.ai ↗
$ how the score was worked out · Meta vendor-reported benchmar… 95%
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 |
|---|---|---|
 Meta vendor-reported benchmark scores for its own Muse models (Muse Wonder model card, Muse Spark 1.3 evaluation methodology) · lab-reported | 4 | 95% |
| prior · pull toward the middle | 5% | |
| weighted average | = 83.8 |
$ 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