Model · Meta
Meta Muse Spark 1.1
3 appearances · meta-score 77.5/100 · last seen 2026-07-19 · scored under rubric v15
Pooled composite across 3 primary runs: 68.3 ± 29.9 (95% CI)
- $1518.02Total cost
- 1,971,829,350Tokens
- 45.6%Cache hit
- 18,202Requests
- 0.02%Failure rate
- 1.22Tools/response
- 1,298,947Tokens per $
Agent role scores
| Agent role | Score / 10 |
|---|---|
| Sprint-Review Agent | 9.04 |
| Develop Agent | 8.56 |
| Code-Review Agent | 8.54 |
| Orchestrator | 8.45 |
| Architect | 8.21 |
| Build Agent | 7.91 |
| Spec Agent | 7.89 |
| Director | 7.69 |
| Test Agent | 7.28 |
| Sprint-Plan Agent | 7.12 |
| Scout Agent | 7.03 |
| Code Agent | 6.99 |
| Platform Agent | 6.64 |
| Pseudocode Agent | 5.82 |
Run history (primary model)
| Run | SoW | Date | Composite |
|---|---|---|---|
| Muse Spark 1.1 | circles | Jul 2026 | 80.8 |
| Muse Spark 1.1 | solarSystem | Jul 2026 | 67.2 |
| Muse Spark 1.1 | the2048 | Jul 2026 | 56.8 |
Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded — no vendor sponsorship, credits, or grants.