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

# Model Deprecations: November 2026

> Models soon leaving Token Factory Serverless with recommended migration paths.

> [!NOTE]
>
> **On November 9, 2026:**
>
> * Affected Serverless models will no longer be available via API or Playground.
> * Dedicated Endpoints for these models will be unaffected by this change.

## What to Know

These models will be removed from Token Factory Serverless on **November 9, 2026**. After that date, you will not be able to select them in the Playground or send API requests using their model IDs.

* This change only affects Serverless usage. [Dedicated Endpoints](https://tokenfactory.nebius.com/endpoints/create-dedicated-endpoint) are unaffected.
* Token Factory does not automatically reroute requests from deprecated models.
* Choose a replacement based on your workload, then test before migrating traffic.

| Deprecating Model | **Recommended Replacement** |
| - | - |
| [`zai-org/GLM-5.1`](https://tokenfactory.nebius.com/models/catalog/text2text/zai-org%2FGLM-5.1) | [`zai-org/GLM-5.2`](https://tokenfactory.nebius.com/models/catalog/text2text/zai-org%2FGLM-5.2) |
| [`nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B`](https://tokenfactory.nebius.com/models/catalog/text2text/nvidia%2FNVIDIA-Nemotron-3-Nano-30B-A3B) | [`nvidia/Nemotron-3_5-Lightning`](https://tokenfactory.nebius.com/models/catalog/text2text/nvidia%2FNemotron-3_5-Lightning) |
| [`nvidia/nemotron-3-super-120b-a12b`](https://tokenfactory.nebius.com/models/catalog/text2text/nvidia%2Fnemotron-3-super-120b-a12b) | [`nvidia/Nemotron-3.5-Super-VL`](https://tokenfactory.nebius.com/models/catalog/image2text/nvidia%2FNemotron-3.5-Super-VL) |
| [`deepseek-ai/DeepSeek-V4-Pro`](https://tokenfactory.nebius.com/models/catalog/text2text/deepseek-ai%2FDeepSeek-V4-Pro) | [`deepseek-ai/DeepSeek-V4-Pro-0813`](https://tokenfactory.nebius.com/models/catalog/text2text/deepseek-ai%2FDeepSeek-V4-Pro-0813) |
| [`deepseek-ai/DeepSeek-V4-Flash-0731`](https://tokenfactory.nebius.com/models/catalog/text2text/deepseek-ai%2FDeepSeek-V4-Flash-0731) | [`deepseek-ai/DeepSeek-V4.1-Flash-1017`](https://tokenfactory.nebius.com/models/catalog/image2text/deepseek-ai%2FDeepSeek-V4.1-Flash-1017) |
| [`deepseek-ai/DeepSeek-V4.1-Flash`](https://tokenfactory.nebius.com/models/catalog/image2text/deepseek-ai%2FDeepSeek-V4.1-Flash) | [`deepseek-ai/DeepSeek-V4.1-Flash-1017`](https://tokenfactory.nebius.com/models/catalog/image2text/deepseek-ai%2FDeepSeek-V4.1-Flash-1017) |
| [`Qwen/Qwen3.5-397B-A17B`](https://tokenfactory.nebius.com/models/catalog/text2text/Qwen%2FQwen3.5-397B-A17B) | [`Qwen/Qwen3.8-27B`](https://tokenfactory.nebius.com/models/catalog/image2text/Qwen%2FQwen3.8-27B) |
| [`Qwen/Qwen3-235B-A22B-Instruct-2507`](https://tokenfactory.nebius.com/models/catalog/text2text/Qwen%2FQwen3-235B-A22B-Instruct-2507) | [`Qwen/Qwen3-30B-A3B-Instruct-2507`](https://tokenfactory.nebius.com/models/catalog/text2text/Qwen%2FQwen3-30B-A3B-Instruct-2507) |
| [`moonshotai/Kimi-K2.6`](https://tokenfactory.nebius.com/models/catalog/image2text/moonshotai%2FKimi-K2.6) | [`moonshotai/Kimi-K3`](https://tokenfactory.nebius.com/models/catalog/image2text/moonshotai%2FKimi-K3) |
| [`moonshotai/Kimi-K2.7-Code`](https://tokenfactory.nebius.com/models/catalog/text2text/moonshotai%2FKimi-K2.7-Code) | [`moonshotai/Kimi-K3`](https://tokenfactory.nebius.com/models/catalog/image2text/moonshotai%2FKimi-K3) |
| [`openbmb/MiniCPM-V-4_5`](https://tokenfactory.nebius.com/models/catalog/image2text/openbmb%2FMiniCPM-V-4_5) | [`nvidia/Nemotron-3.5-Super-VL`](https://tokenfactory.nebius.com/models/catalog/image2text/nvidia%2FNemotron-3.5-Super-VL) |

## Preparing to Migrate

1. Identify applications and workflows that send requests to a deprecating model ID.
2. Compare the replacement model's capabilities, performance and price.
3. Test prompts, inputs and workloads to verify the new model will work for your use cases.
4. Update the `model` and `base_url` values in your API requests, then verify the results.

Refer to our [quick start guide](https://docs.tokenfactory.nebius.com/quickstart.md) for reference and up to date examples.

## Consider a Dedicated Endpoint

For production workloads that need stable model versions or custom configurations, consider [Dedicated Endpoints](https://tokenfactory.nebius.com/endpoints/create-dedicated-endpoint), which provide:

* Control over model versions and deployments.
* Dedicated capacity and configurable scaling.
* Workload isolation.

To learn more about Dedicated Endpoints or for help getting started, contact our team:

👉 [Contact Token Factory Sales](https://nebius.com/services/token-factory/enterprise-grade-inference#token-factory-enterprise-sales-form)

## Need Help?

If you need assistance migrating, choosing the right model or getting started with Dedicated Endpoints, our team is ready to help.

👉 [Contact Token Factory Support](mailto:tokenfactory-support@nebius.com)