Muse Wonder Full
Meta's full-strength build — subscription only, no token pricing
Benchmark scores
How it is measured →Ranked 8 · bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 Intelligence index | 50.0 | 83% | −10.0 pts |
| AA Intelligence Index v4.3.2 Intelligence index (legacy) | 56.0 | 97% | −1.6 pts |
| GPQA Diamond | — | — | not indexed |
| Terminal-Bench 4.0 | — | — | not indexed |
| LiveBench Design | — | — | not indexed |
| Vals Index | — | — | not indexed |
Assessment
AA Intelligence Index v5 50.0 at the max setting, down from 56.0 on v4.3.2 — six points lost is not a regression, it is v5's harder exam (60% of the weight comes from private sets). It takes first overall on GDP.pdf at 33% (all-pass). Available only through the Muse Code subscription ($50 / $200 a month), with no token-priced public API, so it stays out of the price comparison; AA has no output speed or latency figures for it either. The standard Muse Wonder runs on the Meta Model API at $4/$20.
At the same price
This model has no token-priced API, so there is no price tier to compare against.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| No comparable models at this price point | ||||
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard