← All benchmarks
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
NVIDIA · Released 2025-04-07 · Independent benchmark scores from Artificial Analysis
Intelligence Index
9.1
Output speed
53.9
tokens/sec
Blended price
$0.90
per 1M tokens
Time to first token
0.68s
Model metadata
| Model ID | cf095603-72b6-47f8-8ee1-09a42890f92a |
|---|---|
| Slug | llama-3-1-nemotron-ultra-253b-v1-reasoning |
| Release date | 2025-04-07 |
| Provider | NVIDIA |
| Provider slug | nvidia |
| Output speed | 53.9 tok/s |
| Time to first token | 0.68s |
| Time to first answer | 37.81s |
| Blended price ($/1M) | $0.90 |
| Input price ($/1M) | $0.60 |
| Output price ($/1M) | $1.80 |
All evaluation scores (17)
| aime | 74.7% |
|---|---|
| aime 25 | 63.7% |
| artificial analysis coding index | — |
| artificial analysis intelligence index | 9.10 |
| artificial analysis math index | 63.70 |
| gpqa | 72.8% |
| hle | 8.1% |
| ifbench | 38.2% |
| lcr | 7.3% |
| livecodebench | 64.1% |
| math 500 | 95.2% |
| mmlu pro | 82.5% |
| scicode | 34.7% |
| tau banking | — |
| tau2 | 11.4% |
| terminalbench hard | 2.3% |
| terminalbench v2 1 | — |
Source: artificialanalysis.ai/models/llama-3-1-nemotron-ultra-253b-v1-reasoning
Benchmark data from Artificial Analysis.