Benchmark

Meta Llama — reported benchmarks

Meta · source page ↗ · last checked Jul 3, 2026, 12:02 AM

Reported benchmarks · Llama 4 Maverick

captured Jul 3, 2026, 12:02 AM
BenchmarkScore
MMLU Pro80.5%
GPQA Diamond69.8%
LiveCodeBench43.4 pass@1 · averaged over multiple generations
HumanEval86.4%
Multilingual MMLU84.6%
GSM8K95.2%
MATH-50085.3%
SWE-bench Verified74.2% pass@1

Vendor-reported via automated web search — not independently verified. See the cited matrix on /models.

Change history