Benchmark
Epoch AI evaluates Muse Spark 1.3 (xhigh)
FrontierMath-Tiers-1-3-v2-Private 74.4% · FrontierMath-Tier-4-v2-Private 41.5% · Chess Puzzles 35.0% · Mystery Game Puzzles 14.1%
- Type
- BENCHMARK_RESULT
- Detection
- benchmark
- Confidence
- high
- Status
Related entities