← All comparisons
Llama 3.2 Instruct 90B (Vision) vs Claude 4.5 Haiku (Reasoning)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 3.2 Instruct 90B (Vision) | Claude 4.5 Haiku (Reasoning) | |
|---|---|---|
| Intelligence Index | 11.9 | 37.1 |
| Coding Index | — | 32.6 |
| Math Index | — | 83.7 |
| Output speed (tok/s) | 47.1 | 142.2 |
| Blended price ($/1M) | $1.38 | $2.19 |
| Time to first token (s) | 0.48s | 10.48s |
| aime | 5.0% | — |
| aime 25 | — | 83.7% |
| artificial analysis coding index | — | 32.60 |
| artificial analysis intelligence index | 11.90 | 37.10 |
| artificial analysis math index | — | 83.70 |
| gpqa | 43.2% | 67.2% |
| hle | 4.9% | 9.7% |
| ifbench | — | 54.3% |
| lcr | — | 70.3% |
| livecodebench | 21.4% | 61.5% |
| math 500 | 62.9% | — |
| mmlu pro | 67.1% | 76.0% |
| scicode | 24.0% | 43.3% |
| tau2 | — | 54.7% |
| terminalbench hard | — | 27.3% |
Benchmark data from Artificial Analysis.