Model · Meta

Meta Muse Spark 1.2

1 appearance · meta-score 64.8/100 · last seen 2026-08-06 · scored under rubric v15

Pooled composite across 1 primary run: 58.9 - no interval, a single run proves nothing about spread provisional: n < 3, treat as unsettled

$99.16Total cost
161,702,131Tokens
56.8%Cache hit
2,570Requests
0.51%Failure rate
1.07Tools/response
1,630,738Tokens per $

Agent role scores

Per-agent-role scores for Meta Muse Spark 1.2
Agent role	Score / 10
Architect	7.61
Sprint-Review Agent	7.50
Platform Agent	7.01
Code-Review Agent	7.01
Scout Agent	6.33
Code Agent	5.98
Sprint-Plan Agent	5.84
Pseudocode Agent	5.46
Spec Agent	5.44

Run history (primary model)

Runs where Meta Muse Spark 1.2 was the primary model, oldest first
Run	SoW	Date	Composite
Muse Spark 1.2 + Gemini Flash Lite	the2048	Aug 2026	58.9

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.