Llama 3.1 Nemotron Ultra 253B V1
NVIDIA
Key facts · verified Oct 29, 2026
Llama 3.1 Nemotron Ultra 253B V1 is a NVIDIA model released on 7 Apr 2025. Its API price is $0.60 per million input tokens and $1.80 per million output tokens. Its context window is 131.1K tokens. It ranks #225 on LMArena's text leaderboard (Elo 1348, read 2 Nov 2026). Source: huggingface.co.
Llama 3.1 Nemotron Ultra 253B V1 is a 253-billion-parameter model from NVIDIA, released on April 7, 2025 with a 131.1K-token context window and open weights under the NVIDIA Open Model License. Third-party pricing listed in LMArena's dataset is $0.60 per million input tokens and $1.80 per million output tokens. 1348 Elo (rank 225)
Pricing & specs checkedSource: vendor documentation · checked Oct 29, 2026LMArena: 1348 Elo (rank 225) · source · retrieved Nov 2, 2026
Full Specifications
$0.60 / $1.80 per million tokens | |
|---|---|
| Identity | |
| Developer | NVIDIA |
| Released | Apr 2025 |
| Licence | Nvidia Open Model |
| Self-hostable | Yes |
| Cost | |
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $0.900 / 1M tokens |
| Input price | $0.600 / 1M tokens |
| Output price | $1.800 / 1M tokens |
| Capacity | |
| Context window | 131.1K tokens |
| Measured quality | |
| LMArena Elo | 1348 (checked Nov 2026) |
| LMArena rank | Rank 225 |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 1498 |