← All comparisons
Llama 3.2 Instruct 11B (Vision) vs Claude 3.5 Sonnet (Oct '24)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 3.2 Instruct 11B (Vision) | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 8.7 | 15.9 |
| Coding Index | 4.2 | 30.2 |
| Math Index | 1.7 | — |
| Output speed (tok/s) | 77.0 | 0.0 |
| Blended price ($/1M) | $0.24 | $6.56 |
| Time to first token (s) | 0.49s | 0.00s |
| aime | 9.3% | 15.7% |
| aime 25 | 1.7% | — |
| artificial analysis coding index | 4.20 | 30.20 |
| artificial analysis intelligence index | 8.70 | 15.90 |
| artificial analysis math index | 1.70 | — |
| gpqa | 22.1% | 59.9% |
| hle | 5.2% | 3.9% |
| ifbench | 30.4% | — |
| lcr | 11.7% | — |
| livecodebench | 11.0% | 38.1% |
| math 500 | 51.6% | 77.1% |
| mmlu pro | 46.4% | 77.2% |
| scicode | 11.2% | 36.6% |
| tau2 | 14.6% | — |
| terminalbench hard | 0.8% | — |
Benchmark data from Artificial Analysis.