← All comparisons
Llama 4 Scout vs Claude 3.5 Sonnet (Oct '24)
Meta vs Anthropic — side-by-side benchmark comparison
| Llama 4 Scout | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 13.5 | 15.9 |
| Coding Index | 6.7 | 30.2 |
| Math Index | 14.0 | — |
| Output speed (tok/s) | 113.8 | 0.0 |
| Blended price ($/1M) | $0.29 | $6.56 |
| Time to first token (s) | 0.56s | 0.00s |
| aime | 28.3% | 15.7% |
| aime 25 | 14.0% | — |
| artificial analysis coding index | 6.70 | 30.20 |
| artificial analysis intelligence index | 13.50 | 15.90 |
| artificial analysis math index | 14.00 | — |
| gpqa | 58.7% | 59.9% |
| hle | 4.3% | 3.9% |
| ifbench | 39.5% | — |
| lcr | 25.8% | — |
| livecodebench | 29.9% | 38.1% |
| math 500 | 84.4% | 77.1% |
| mmlu pro | 75.2% | 77.2% |
| scicode | 17.0% | 36.6% |
| tau2 | 15.5% | — |
| terminalbench hard | 1.5% | — |
Benchmark data from Artificial Analysis.