Claude 3.5 Sonnet (Anthropic) and Llama Nemotron Ultra 253B (NVIDIA) are both frontier AI models. Claude 3.5 Sonnet was published in June 2024 and Llama Nemotron Ultra 253B in March 2025.
Llama Nemotron Ultra 253B was trained on 3.9 x 10^25 FLOP, about 1.4x the compute of Claude 3.5 Sonnet at 2.7 x 10^25 FLOP. Training compute is the closest available proxy for how much was invested in a model, though it says nothing on its own about how well that compute was spent.
These two models share no benchmark on which both have been scored, so no direct performance comparison is possible here. The specification table below is a comparison of inputs, not of results.
| Field | Claude 3.5 Sonnet | Llama Nemotron Ultra 253B |
|---|---|---|
| Organization | Anthropic | NVIDIA |
| Published | Jun 20, 2024 | Mar 18, 2025 |
| Training compute | 2.7 x 10^25 FLOP | 3.9 x 10^25 FLOP |
| Parameters | — | 253B |
| Dataset size | — | 603B |
| Training cost (2023 USD) | $26M | — |
| Accessibility | API access | Open weights (restricted use) |
| Open weights | No | Yes |
| Country | United States | United States |
