← All comparisons
Llama 3.2 Instruct 90B (Vision) vs Claude 4.1 Opus (Non-reasoning)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 3.2 Instruct 90B (Vision) | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 11.9 | 36.0 |
| Coding Index | — | — |
| Math Index | — | — |
| Output speed (tok/s) | 47.1 | 44.7 |
| Blended price ($/1M) | $1.38 | $32.81 |
| Time to first token (s) | 0.48s | 1.63s |
| aime | 5.0% | — |
| aime 25 | — | — |
| artificial analysis coding index | — | — |
| artificial analysis intelligence index | 11.90 | 36.00 |
| artificial analysis math index | — | — |
| gpqa | 43.2% | — |
| hle | 4.9% | — |
| ifbench | — | — |
| lcr | — | — |
| livecodebench | 21.4% | — |
| math 500 | 62.9% | — |
| mmlu pro | 67.1% | — |
| scicode | 24.0% | — |
| tau2 | — | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.