← All comparisons
Llama 4 Maverick vs Claude 4.1 Opus (Non-reasoning)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 4 Maverick | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 18.4 | 36.0 |
| Coding Index | 15.6 | — |
| Math Index | 19.3 | — |
| Output speed (tok/s) | 119.7 | 44.7 |
| Blended price ($/1M) | $0.47 | $32.81 |
| Time to first token (s) | 0.64s | 1.63s |
| aime | 39.0% | — |
| aime 25 | 19.3% | — |
| artificial analysis coding index | 15.60 | — |
| artificial analysis intelligence index | 18.40 | 36.00 |
| artificial analysis math index | 19.30 | — |
| gpqa | 67.1% | — |
| hle | 4.8% | — |
| ifbench | 43.0% | — |
| lcr | 46.0% | — |
| livecodebench | 39.7% | — |
| math 500 | 88.9% | — |
| mmlu pro | 80.9% | — |
| scicode | 33.1% | — |
| tau2 | 17.8% | — |
| terminalbench hard | 6.8% | — |
Benchmark data from Artificial Analysis.