# deepseek-v4-flash

Calls to the model deepseek-v4-flash on Azure AI Foundry stop working on February 20, 2028.

On February 20, 2028, model deepseek-v4-flash will be retired on Azure AI Foundry. Any calls directed to model deepseek-v4-flash will fail after this date. This retirement impacts workloads utilizing this specific model.

## What to use instead

The listed replacement deepseek-v4-flash-0731 stops working on December 3, 2026. No replacement that holds is named yet.

- API: Azure AI Foundry (*.services.ai.azure.com)
- Model: deepseek-v4-flash
- Released: April 24, 2026 (models.dev)
- Kind: Model retirement
- Takes effect: February 20, 2028
- Calls after that day: fail
- Calls it hits: `model deepseek-v4-flash`
- Source: [Deprecations page](https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule)
- Verified: November 6, 2026, against Azure AI Foundry's own pages
- Page: https://www.okapi.bot/models/azure-ai-foundry/deepseek-v4-flash/

## Azure AI Foundry's words

> Model: DeepSeek-V4-Flash; Version: 2026-04-23; Lifecycle: GA; Retirement date: 2028-02-20; Replacement: DeepSeek-V4-Flash-0731

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