#28
Muse Spark
Meta · Proprietary · Reasoning · 262K context
57/ 100 composite
Released 2026-04-08
Benchmark Scores by Category
Coding
45.2
Knowledge
74.6
Reasoning
52.8
Agentic
47.9
Math
56.2
Multimodal
77.3
Scores based on weighted benchmarks: SWE-bench, LiveCodeBench (Coding); HLE, MMLU-Pro (Knowledge); ARC-AGI-2, LongBench v2 (Reasoning); Terminal-Bench, BrowseComp (Agentic); FrontierMath, AIME (Math).
API Pricing
API pricing not publicly available for this model.
Estimate Your Costs
Use our cost calculators to estimate monthly spend based on your specific workload: