Muse Spark Open
Muse Spark Open is an 816B total / 42B active open-weight multimodal reasoning model from Meta for coding and long-horizon agentic work, with 1M-token context.
Shogo Agent Index
UnrankedNot ranked yet: 1 agentic benchmark measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.
Agentic coding
54
estimated
Long-horizon autonomy
58
estimated
Knowledge-work agents
35
32% measured
Computer use
37
estimated
Security & context learning
49
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 78 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 74 | |
| DeepSWE | est. | 6% | predicted 50 | |
| FrontierSWE | est. | 5% | predicted 17 | |
| CursorBench | est. | 4% | predicted 38 | |
| GSO-Bench | est. | 4% | predicted 31 | |
| METR Time Horizons | est. | 10% | predicted 76 | |
| Vending-Bench 2 | est. | 6% | predicted 51 | |
| PostTrainBench | est. | 4% | predicted 26 | |
| APEX-Agents | 33.4% | 8% | source | |
| GDPval | est. | 7% | predicted 48 | |
| Remote Labor Index | est. | 5% | predicted 4 | |
| DeepResearch Bench | est. | 5% | predicted 48 | |
| OSWorld 2.0 | est. | 6% | predicted 10 | |
| OSWorld | est. | 4% | predicted 78 | |
| Cybench | est. | 5% | predicted 76 | |
| CL-bench | est. | 5% | predicted 22 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| ARC-AGI | Reasoning | 89.0% | source |
| ARC-AGI-2 | Reasoning | 57.5% | source |
| CritPt | Reasoning | 14.9% | source |
| FrontierMath Tiers 1-3 | Math | 69.8% | source |
| FrontierMath Tier 4 | Math | 31.7% | source |
| GPQA Diamond | Reasoning | 88.6% | source |
| OTIS Mock AIME | Math | 98.1% | source |
| SciCode | Coding | 58.4% | source |
| SimpleQA Verified | Knowledge | 56.6% | source |
API hosts
Live prices and uptime for every host serving Muse Spark Open via OpenRouter.