> ## Documentation Index
> Fetch the complete documentation index at: https://docs.novita.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Nov 26, 2026 Model Deprecation Notice

> Novita retires 13 models, including google/gemma-3-27b-it, meta-llama/llama-3.3-70b-instruct and openai/gpt-oss-120b, at 16:00 UTC on Nov 26, 2026. No replacement models are provided; stop using the affected models before the cutoff.

The following models will be retired from the Novita platform at **16:00 UTC on November 26, 2026**. No replacement models are provided for this retirement. If your application uses an affected model, please update it to stop using the model before the retirement time.

## Affected models

| Model ID | Retirement time | Replacement model |
| - | - | - |
| `Sao10K/L3-8B-Stheno-v3.2` | Nov 26, 2026, 16:00 UTC | — |
| `google/gemma-3-27b-it` | Nov 26, 2026, 16:00 UTC | — |
| `google/gemma-4-26b-a4b-it` | Nov 26, 2026, 16:00 UTC | — |
| `google/gemma-4-31b-it` | Nov 26, 2026, 16:00 UTC | — |
| `meta-llama/llama-3.1-8b-instruct` | Nov 26, 2026, 16:00 UTC | — |
| `meta-llama/llama-3.3-70b-instruct` | Nov 26, 2026, 16:00 UTC | — |
| `openai/gpt-oss-120b` | Nov 26, 2026, 16:00 UTC | — |
| `zai-org/autoglm-phone-9b-multilingual` | Nov 26, 2026, 16:00 UTC | — |
| `nvidia/nemotron-3-nano-30b-a3b` | Nov 26, 2026, 16:00 UTC | — |
| `deepseek/deepseek-ocr-2` | Nov 26, 2026, 16:00 UTC | — |
| `baai/bge-reranker-v2-m3` | Nov 26, 2026, 16:00 UTC | — |
| `qwen/qwen3-omni-30b-a3b-thinking` | Nov 26, 2026, 16:00 UTC | — |
| `minimaxai/minimax-m1-80k` | Nov 26, 2026, 16:00 UTC | — |

## Action required

After the retirement time, API requests to these models will return an error. Please update your application to stop using the affected models before the cutoff. If you need assistance, reach out to us via our **Discord** community or submit a support ticket.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.