← All comparisons
Gemma 3 12B Instruct vs Claude 3.5 Sonnet (June '24)
Google vs Anthropic — side-by-side benchmark comparison
| Gemma 3 12B Instruct | Claude 3.5 Sonnet (June '24) | |
|---|---|---|
| Intelligence Index | 8.8 | 14.2 |
| Coding Index | 6.3 | 26.0 |
| Math Index | 18.3 | — |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.14 | $6.56 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | 22.0% | 9.7% |
| aime 25 | 18.3% | — |
| artificial analysis coding index | 6.30 | 26.00 |
| artificial analysis intelligence index | 8.80 | 14.20 |
| artificial analysis math index | 18.30 | — |
| gpqa | 34.9% | 56.0% |
| hle | 4.8% | 3.7% |
| ifbench | 36.7% | — |
| lcr | 6.7% | — |
| livecodebench | 13.7% | — |
| math 500 | 85.3% | 69.5% |
| mmlu pro | 59.5% | 75.1% |
| scicode | 17.4% | 31.6% |
| tau2 | 10.8% | — |
| terminalbench hard | 0.8% | — |
Benchmark data from Artificial Analysis.