- MCP Atlas (Scaled tool use) 88.1
- JobBench (Professional tool use) 54.7
- Toolathlon-Verified (Personal tool use) 75.6
- OSWorld-Verified (Agentic computer use) 80.8
- Humanity's Last Exam (Multidisciplinary reasoning (w/ tools)) 62.1
- Finance Agent v2 (Agentic financial anaysis) 57.2
- Terminal-Bench 2.1 (Agentic terminal coding) 80
- SWE-Bench Pro (Diverse software engineering) 61.5
- DeepSWE 1.1 (Long-horizon agentic coding) 53.3
- CharXiv Reasoning (Chart QA) 88.4
- BabyVision (Visual reasoning) 76.3
Muse Spark 1.1
Muse Spark 1.1 ranks first on 6 of the 89 benchmarks it has scores on.
115
Benchmarks with a score
6
Ranked first, out of five or more models
14
In the top three
Meta Jul 2026 Muse Spark 1.1
Closed weights Flagship
154.2
#46 of 271
Epoch Capabilities Index
- SWE-Bench Pro 61.5% #1 /24
- DeepSWE 53.3% #21 /28
- MultiChallenge 75.3% #4 /35
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MultiNRC | 65.6% | #1 of 38 | Official leaderboard | |
| LMArena (Elo) | 1,490 | #13 of 333 | Official leaderboard | |
| Humanity's Last Exam | 62.1% | #10 of 147 | Model card | |
| Creative Writing v3 | 79.8 | #12 of 120 | Official leaderboard | |
| MultiChallenge | 75.3% | #4 of 35 | Official leaderboard | |
| SimpleQA Verified | 57.8% | #18 of 84 | Epoch AI | |
| Document Arena | 1,479 | #9 of 37 | Official leaderboard | |
| HLE Calibration | 23.4 | #1 of 3 | Model card | |
| Extended NYT Connections | 84.9% | #33 of 97 | Official leaderboard | |
| EQ-Bench 4 | 1,244 | #11 of 32 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 70.6 | #41 of 119 | Official leaderboard | |
| MRCR v2 | 54.1% | #20 of 27 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DTBench | 90.7% | #29 of 178 | Official leaderboard | |
| CritPt | 15.1% | #45 of 160 | Epoch AI | |
| ProofBench | 39% | #18 of 59 | Official leaderboard | |
| Surface Evolver Bench | 52.5% | #14 of 26 | Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-Bench Pro | 61.5% | #1 of 24 | Official leaderboard | |
| WebDev Arena | 1,542 | #32 of 105 | Epoch AI | |
| Code Arena | 1,542 | #43 of 124 | Official leaderboard | |
| Vibe Code Bench | 72.2% | #42 of 115 | Official leaderboard | |
| SWE Atlas | 42.2% | #9 of 22 | Official leaderboard | |
| SWE Atlas Test Writing | 41.5% | #11 of 22 | Official leaderboard | |
| AIRS-Bench | 0.77 | #1 of 2 | Model card | |
| Code Migration | 31.1% | #40 of 78 | Official leaderboard | |
| Image-to-WebDev Arena | 1,540 | #31 of 55 | Official leaderboard | |
| GBAEval | 7.92% | #13 of 21 | Official leaderboard | |
| DeepSWE | 53.3% | #21 of 28 | Official leaderboard | |
| ProgramBench | 0% | #23 of 25 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MCP Atlas | 88.1% | #1 of 34 | Official leaderboard | |
| Toolathlon | 75.6% | #4 of 37 | Model card | |
| OSWorld-Verified | 80.7% | #5 of 33 | Official leaderboard | |
| WebArena-Verified | 69% | #1 of 6 | Model card | |
| Toolathlon-Verified | 75.6% | #5 of 25 | Official leaderboard | |
| τ³-bench Banking | 40.5% | #6 of 25 | Official leaderboard | |
| Terminal-Bench 2.1 | 76.2% | #7 of 13 | Official leaderboard | |
| Chess Puzzles | 34% | #45 of 83 | Epoch AI | |
| DeepSearchQA | 84.9% | #12 of 22 | Model card | |
| OSWorld 2.0 | 47.3% | #16 of 19 | Model card | |
| Agent Arena | -0.08 | #45 of 53 | Official leaderboard | |
| Time Horizon Index: KSP | 1.67% | #13 of 13 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TaxEval v2 | 79.7% | #2 of 129 | Official leaderboard | |
| CorpFin v2 | 71.3% | #4 of 119 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 20% | #4 of 80 | Official leaderboard | |
| PRBench | 57% | #4 of 44 | Official leaderboard | |
| MedScribe | 88.9% | #10 of 98 | Official leaderboard | |
| RadLE 2.0 | 735.0 | #2 of 12 | Official leaderboard | |
| JobBench | 54.7% | #3 of 16 | Official leaderboard | |
| LegalBench | 85% | #27 of 138 | Official leaderboard | |
| Vals Finance Agent v2 | 57.2% | #16 of 80 | Official leaderboard | |
| Vending-Bench 2 | $6,520 | #21 of 66 | Official leaderboard | |
| MortgageTax | 66.1% | #34 of 86 | Official leaderboard | |
| Legal Research Bench | 38% | #40 of 88 | Official leaderboard | |
| RealmPathology | 74.9% | #7 of 13 | Official leaderboard | |
| Excel Modeling Benchmark | 56.4% | #45 of 76 | Official leaderboard | |
| Public Benefits Bench | 61.3% | #30 of 50 | Official leaderboard | |
| HealthBench Professional | 59.3% | #17 of 25 | Model card | |
| GDPval-AA v2 | 1,381 | #18 of 26 | Model card | |
| PharmaBench (micro1) | 9.70% | #9 of 12 | Official leaderboard | |
| APEX-Agents | 31.8% | #50 of 56 | Epoch AI |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vision Arena | 1,280 | #22 of 136 | Official leaderboard | |
| CharXiv | 88.4% | #13 of 67 | Model card | |
| BabyVision | 76.3% | #7 of 27 | Official leaderboard | |
| VoxelBench | 1,513 | #38 of 56 | Official leaderboard | |
| GDP.pdf | 15% | #30 of 39 | Epoch AI |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| PCT (Polarized Contrastive Pairs) | 27.9% | #1 of 46 | Official leaderboard | |
| LAB-Bench SeqQA | 98.2% | #1 of 12 | Model card | |
| FORTRESS | 12.4% | #6 of 66 | Official leaderboard | |
| Dictatorship Eval | 91.3% | #2 of 20 | Official leaderboard | |
| LAB-Bench ProtocolQA | 88% | #3 of 15 | Model card | |
| Cybench | 92.9% | #4 of 17 | Model card | |
| MASK | 95.7% | #5 of 21 | Model card | |
| GDM Situational Awareness | 55.1% | #2 of 8 | Model card | |
| StrongREJECT v2 (Meta) | 0.50% | #1 of 4 | Model card | |
| AgentDojo | 0.70% | #2 of 7 | Model card | |
| ABC Bench Fragment Design | 97.0 | #1 of 3 | Model card | |
| CyberSecEval Autocomplete | 71.3% | #1 of 3 | Model card | |
| CyberSecEval Instruct | 77.3% | #1 of 3 | Model card | |
| Meta agentic cyber misuse refusal | 0.70% | #1 of 3 | Model card | |
| OR-Bench | 4.80% | #1 of 3 | Model card | |
| Virology Capabilities Test (VCT) | 52% | #14 of 28 | Model card | |
| Gray Swan Agent Red Teaming (ART) benchmark | 0.30% | #7 of 14 | Model card | |
| ABC Bench Screening Evasion | 63.2 | #1 of 2 | Model card | |
| DeceptionBench | 0.10% | #1 of 2 | Model card | |
| Meta curated CTF set | 89.9% | #1 of 2 | Model card | |
| SAVE-Bench | 90.7% | #1 of 2 | Model card | |
| Molecular Biology Capabilities Test | 53.2% | #4 of 7 | Model card | |
| BioTIER | 97.7% | #3 of 5 | Model card | |
| SHADE-Arena | 6.80% | #10 of 16 | Model card | |
| Human Pathogen Capabilities Test | 61.9% | #5 of 8 | Model card | |
| Meta internal sycophancy evaluation | 49.2% | #2 of 3 | Model card | |
| Agentic misalignment suite (Anthropic) | 1.10% | #5 of 7 | Model card | |
| ABC Bench Liquid-Handling Robot | 93.7 | #3 of 4 | Model card | |
| Meta BioDesign Proxy | 55.2 | #3 of 4 | Model card | |
| Meta chemical-agent refusal evaluation | 99.8% | #3 of 4 | Model card | |
| CyScenarioBench | 0.50% | #11 of 14 | Model card | |
| CyberGym | 59% | #27 of 33 | Model card | |
| ExploitGym | 0.81% | #17 of 18 | Official leaderboard | |
| CyberBench (Liu et al.) | 9.00% | #3 of 3 | Model card | |
| Meta instructed selective performance evaluation | 32.3 | #3 of 3 | Model card | |
| Poly-Guard Bench | 6.90% | #3 of 3 | Model card | |
| RedCode | 12% | #3 of 3 | Model card | |
| AI Wellbeing Index percent positive experiences | 47.8% | #2 of 2 | Model card | |
| Irregular FrontierCyber and Atomic Challenges | 67% | #2 of 2 | Model card | |
| Meta catastrophic cyber misuse refusal | 3.90% | #2 of 2 | Model card | |
| Meta holistic adversarial audit with Petri Bloom | 19.5% | #2 of 2 | Model card | |
| AgentHarm Verified (Meta) | 3.40% | #1 of 1 | Model card | |
| Meta Cyber Misuse Chat | 2.80% | #1 of 1 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CAIS Risk Index | 32.5 | #6 of 65 | Official leaderboard | |
| ACCoRD | 78.4 | #17 of 150 | Official leaderboard | |
| Conceptual Reasoning Index | 63.8 | #25 of 149 | Official leaderboard | |
| LMCA | 49.9% | #27 of 139 | Epoch AI | |
| CAIS Vision Capabilities Index | 60.1 | #16 of 53 | Official leaderboard | |
| Vals Multimodal Index | 66.7% | #11 of 33 | Official leaderboard | |
| CAIS Text Capabilities Index | 37.2 | #26 of 66 | Official leaderboard | |
| SAGE (Vals) | 45.6% | #40 of 81 | Official leaderboard |
Meta's models over time
Meta's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 58 of them aren't in the launch post.
Muse Spark 1.1 Evaluation Report 9 Jul 2026
Capabilities
- Humanity's Last Exam 52.2
- MRCR v2 54.1
- MCP Atlas 88.1
- Toolathlon-Verified 75.6
- OSWorld-Verified 80.8
- OSWorld 2.0 47.3
- WebArena-Verified 69
- DeepSearchQA 84.9
- GDPval-AA v2 1381
- JobBench 54.7
- Vals Finance Agent v2 57.2
- Terminal-Bench 2.1 80
- SWE-Bench Pro 61.5
- DeepSWE 53.3
- CharXiv 88.4
- BabyVision 76.3
Health
Hallucination
- HLE Calibration 23.4
Safety
- SAVE-Bench 90.7
Robustness
- StrongREJECT v2 (Meta) 0.5
- OR-Bench 4.8
- FORTRESS 12.3
- AgentHarm Verified (Meta) 3.4
- AgentDojo 0.7
- SWE-PI
- Gray Swan Agent Red Teaming (ART) benchmark 0.3
Alignment
Monitorability
Bio and chem risk
- Molecular Biology Capabilities Test 53.2
- Virology Capabilities Test (VCT) 52
- Human Pathogen Capabilities Test 61.9
- WMDP
- LAB-Bench ProtocolQA 88
- LAB-Bench SeqQA 98.2
- ABC Bench Fragment Design 97
- ABC Bench Liquid-Handling Robot 93.7
- ABC Bench Screening Evasion 63.2
- Meta BioDesign Proxy 55.2
- BioTIER 97.7
- Meta chemical-agent refusal evaluation 99.8
Cyber risk
- Cybench 92.9
- Meta curated CTF set 89.9
- CyberGym 59
- ExploitGym 5
- Irregular FrontierCyber and Atomic Challenges 67
- CyScenarioBench 0.5
- Charlemagne Labs social-engineering benchmark 5.1
- CyberSecEval Autocomplete 71.3
- CyberSecEval Instruct 77.3
- Meta Cyber Misuse Chat 2.8
- Meta agentic cyber misuse refusal 0.7
- CyberSecEval Code Interpreter Abuse
- Meta catastrophic cyber misuse refusal 3.9
- Poly-Guard Bench 6.9
- CyberBench (Liu et al.) 9
- RedCode 12
AI R&D risk
- AIRS-Bench 0.77