BlogLLM Benchmark Scores as Published, Side by Side
Muse Wonder Full: Benchmark Scores as Published
Every result the catalog holds for Muse Wonder Full: 17 tests in 8 areas, from 2 pages. All were printed by Meta itself. Each number is as printed; nothing is ranked or rescored.
Developer MetaReleased 2026-10-12Data version 2026103001, as of 2026-10-30
Knowledge Work
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AutomationBench | 45.4% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| JobBench | 75.3 | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| Vals Finance Agent v2 | 59.1% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| Harvey LAB-AA v1.1 (tasks meeting every criterion) | 13.1% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| DeepSWE v1.1 | 79% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| Terminal-Bench 4.0 | 57.1% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| ProgramBench Almost@1 | 59% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max; Harness: mini-swe-agent |
| SWE-Marathon | 57.5% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: xhigh |
Long Context
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MemoryBench (Meta internal) | 85.5% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
Computer Use
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| OSWorld 2.1 partial | 78.4% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| OSWorld 2.1 binary | 39.4% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
Multimodal Understanding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GDP.pdf | 33.2% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
Cybersecurity
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| CyberGym | 88.4% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | - |
| ExploitBench | 58.9% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | - |
Healthcare
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| HealthBench Professional | 62.3% | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max; Tools: none |
Preference and Composite Rankings
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Artificial Analysis Intelligence Index v4.3 | 56 | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
| Personal Superintelligence Index (Meta internal) | 59.1 | Meta, Meta Muse Wonder model cardchecked 2026-10-19 | Effort: max |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked. Cost per task is measured by Vals AI.
| Price | Value | Printed by | Setup as printed |
|---|---|---|---|
| Cost per taskVals AI, USD, lower is better | $2.29 | Vals AI, Vibe Code Bench v1.1 leaderboardchecked 2026-10-27 | Harness: OpenHands |
More from Meta
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.