# nvidia/llama-3_1-nemotron-ultra-253b-v1

Calls to the model nvidia/llama-3_1-nemotron-ultra-253b-v1 on Nebius stopped working on August 31, 2026.

On August 31, 2026, Nebius will retire the model nvidia/llama-3_1-nemotron-ultra-253b-v1. Any API requests targeting model nvidia/llama-3_1-nemotron-ultra-253b-v1 will fail after this date.

## What to use instead

Update requests to use nvidia/nemotron-3-super-120b-a12b, which is recommended by Nebius.

- API: Nebius (api.studio.nebius.ai)
- Model: nvidia/llama-3_1-nemotron-ultra-253b-v1
- Kind: Model retirement
- Takes effect: August 31, 2026
- Calls after that day: fail
- Calls it hits: `model nvidia/llama-3_1-nemotron-ultra-253b-v1`
- Source: [Model notices](https://docs.tokenfactory.nebius.com/llms.txt#deprecation)
- Verified: October 26, 2026, against Nebius's own pages
- Page: https://www.okapi.bot/models/nebius/nvidia/llama-3_1-nemotron-ultra-253b-v1/

Okapi finds the calls this change hits in your repo and opens the pull request with the fix: https://github.com/apps/okapi-agent