← All comparisons
Olmo 3 7B Instruct vs Claude 4 Sonnet (Reasoning)
Allen Institute for AI vs Anthropic — side-by-side benchmark comparison
| Olmo 3 7B Instruct | Claude 4 Sonnet (Reasoning) | |
|---|---|---|
| Intelligence Index | 8.1 | 38.7 |
| Coding Index | 3.4 | 34.1 |
| Math Index | 41.3 | 74.3 |
| Output speed (tok/s) | 0.0 | 55.5 |
| Blended price ($/1M) | $0.13 | $6.56 |
| Time to first token (s) | 0.00s | 8.92s |
| aime | — | 77.3% |
| aime 25 | 41.3% | 74.3% |
| artificial analysis coding index | 3.40 | 34.10 |
| artificial analysis intelligence index | 8.10 | 38.70 |
| artificial analysis math index | 41.30 | 74.30 |
| gpqa | 40.0% | 77.7% |
| hle | 5.8% | 9.6% |
| ifbench | 32.8% | 54.7% |
| lcr | 0.0% | 64.7% |
| livecodebench | 26.6% | 65.5% |
| math 500 | — | 99.1% |
| mmlu pro | 52.2% | 84.2% |
| scicode | 10.3% | 40.0% |
| tau2 | 12.6% | 64.6% |
| terminalbench hard | 0.0% | 31.1% |
Benchmark data from Artificial Analysis.