← All comparisons
ERNIE 4.5 300B A47B vs Claude 4.1 Opus (Non-reasoning)
Baidu vs Anthropic — side-by-side benchmark comparison
| ERNIE 4.5 300B A47B | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 15.0 | 36.0 |
| Coding Index | 14.5 | — |
| Math Index | 41.3 | — |
| Output speed (tok/s) | 25.7 | 44.7 |
| Blended price ($/1M) | $0.48 | $32.81 |
| Time to first token (s) | 1.68s | 1.63s |
| aime | 49.3% | — |
| aime 25 | 41.3% | — |
| artificial analysis coding index | 14.50 | — |
| artificial analysis intelligence index | 15.00 | 36.00 |
| artificial analysis math index | 41.30 | — |
| gpqa | 81.1% | — |
| hle | 3.5% | — |
| ifbench | 39.1% | — |
| lcr | 2.3% | — |
| livecodebench | 46.7% | — |
| math 500 | 93.1% | — |
| mmlu pro | 77.6% | — |
| scicode | 31.5% | — |
| tau2 | 0.0% | — |
| terminalbench hard | 6.1% | — |
Benchmark data from Artificial Analysis.