← All comparisons
ERNIE 4.5 300B A47B vs Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)
Baidu vs NVIDIA — side-by-side benchmark comparison
| ERNIE 4.5 300B A47B | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) | |
|---|---|---|
| Intelligence Index | 15.0 | 14.4 |
| Coding Index | 14.5 | — |
| Math Index | 41.3 | 50.0 |
| Output speed (tok/s) | 25.7 | 0.0 |
| Blended price ($/1M) | $0.48 | $0.00 |
| Time to first token (s) | 1.68s | 0.00s |
| aime | 49.3% | 70.7% |
| aime 25 | 41.3% | 50.0% |
| artificial analysis coding index | 14.50 | — |
| artificial analysis intelligence index | 15.00 | 14.40 |
| artificial analysis math index | 41.30 | 50.00 |
| gpqa | 81.1% | 40.8% |
| hle | 3.5% | 5.1% |
| ifbench | 39.1% | 25.5% |
| lcr | 2.3% | 0.0% |
| livecodebench | 46.7% | 49.3% |
| math 500 | 93.1% | 94.7% |
| mmlu pro | 77.6% | 55.6% |
| scicode | 31.5% | 10.1% |
| tau2 | 0.0% | 11.7% |
| terminalbench hard | 6.1% | — |
Benchmark data from Artificial Analysis.