← All comparisons
Llama 3.2 Instruct 3B vs Claude 4.1 Opus (Non-reasoning)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 3.2 Instruct 3B | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 9.7 | 36.0 |
| Coding Index | — | — |
| Math Index | 3.3 | — |
| Output speed (tok/s) | 52.7 | 44.7 |
| Blended price ($/1M) | $0.15 | $32.81 |
| Time to first token (s) | 0.65s | 1.63s |
| aime | 6.7% | — |
| aime 25 | 3.3% | — |
| artificial analysis coding index | — | — |
| artificial analysis intelligence index | 9.70 | 36.00 |
| artificial analysis math index | 3.30 | — |
| gpqa | 25.5% | — |
| hle | 5.2% | — |
| ifbench | 26.2% | — |
| lcr | 2.0% | — |
| livecodebench | 8.3% | — |
| math 500 | 48.9% | — |
| mmlu pro | 34.7% | — |
| scicode | 5.2% | — |
| tau2 | 21.1% | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.