# Qwen Qwen3 VL 8B Instruct — deprecation status

> Yes — Qwen deprecated Qwen3 VL 8B Instruct and shuts it down on October 10, 2026 (in 68 days). The recommended replacement is Qwen3.6 Flash.

The small instruction-tuned vision model, 8B and open-weighted, which made it the usual choice for self-hosted image understanding in the Qwen ecosystem. The hosted id goes on 10 October 2026; the weights do not.

## Facts

| Field | Value |
| --- | --- |
| Provider | Qwen |
| Model id | `qwen3-vl-8b-instruct` |
| Status | Deprecated |
| Shutdown | October 10, 2026 (in 68 days) |
| Last verified | August 3, 2026 |

## Replacement

- **Qwen3.6 Flash** (`qwen3.6-flash`) · recommended — https://modeldeprecations.dev/qwen/qwen3.6-flash
  - Parameters it accepts: https://modelparams.dev/models/qwen/qwen3.6-flash

```diff
- "model": "qwen3-vl-8b-instruct"
+ "model": "qwen3.6-flash"
```

## FAQ

### When will Qwen Qwen3 VL 8B Instruct shut down?

Qwen Qwen3 VL 8B Instruct is scheduled to shut down on October 10, 2026, in 68 days.

### What replaces Qwen Qwen3 VL 8B Instruct?

Qwen recommends Qwen3.6 Flash (`qwen3.6-flash`) as the replacement for Qwen Qwen3 VL 8B Instruct.

### Is Qwen Qwen3 VL 8B Instruct still available?

Yes, for now. Qwen Qwen3 VL 8B Instruct is deprecated but still answers requests until October 10, 2026. Deprecated models get no further updates and are the first to lose capacity.

### How do I migrate from Qwen Qwen3 VL 8B Instruct to qwen3.6-flash?

Swap the model id in your API call: replace "qwen3-vl-8b-instruct" with "qwen3.6-flash". Keep the rest of the request the same, then re-check the parameters the new model accepts before you ship.

## Sources

- [Alibaba Cloud Model Studio model decommissioning](https://www.alibabacloud.com/help/en/model-studio/model-depreciation) — accessed 2026-08-03

---

Canonical page: https://modeldeprecations.dev/qwen/qwen3-vl-8b-instruct
JSON: https://modeldeprecations.dev/api/v1/models/qwen/qwen3-vl-8b-instruct.json