# OpenAI GPT-3.5 Turbo — deprecation status

> Yes — OpenAI deprecated GPT-3.5 Turbo on April 22, 2026 and shuts it down on October 23, 2026 (in 85 days). The recommended replacement is GPT-5.6 Terra.

The model that made the chat completions API a commodity: cheap enough to put behind a free product, good enough that most people stopped writing prompt chains. It survived every deprecation wave from 2023 to 2026 because so much code was written against it — and it is finally going in October 2026.

## Facts

| Field | Value |
| --- | --- |
| Provider | OpenAI |
| Model id | `gpt-3.5-turbo` |
| Status | Deprecated |
| Aliases | `gpt-3.5-turbo-0125`, `gpt-3.5-turbo-completions` |
| Released | March 1, 2023 |
| Deprecated | April 22, 2026 |
| Shutdown | October 23, 2026 (in 85 days) |
| Last verified | July 30, 2026 |

## Replacement

- **GPT-5.6 Terra** (`gpt-5.6-terra`) · recommended — https://modeldeprecations.dev/openai/gpt-5.6-terra
  - The biggest gotcha is max_tokens: current models use max_completion_tokens and reject the old parameter. Prompts tuned for GPT-3.5 are usually far more verbose than a 5.6-generation model needs.
  - Parameters it accepts: https://modelparams.dev/models/openai/gpt-5.6-terra

```diff
- "model": "gpt-3.5-turbo"
+ "model": "gpt-5.6-terra"
```

## FAQ

### When will OpenAI GPT-3.5 Turbo shut down?

OpenAI GPT-3.5 Turbo is scheduled to shut down on October 23, 2026, in 85 days.

### What replaces OpenAI GPT-3.5 Turbo?

OpenAI recommends GPT-5.6 Terra (`gpt-5.6-terra`) as the replacement for OpenAI GPT-3.5 Turbo. The biggest gotcha is max_tokens: current models use max_completion_tokens and reject the old parameter. Prompts tuned for GPT-3.5 are usually far more verbose than a 5.6-generation model needs.

### Is OpenAI GPT-3.5 Turbo still available?

Yes, for now. OpenAI GPT-3.5 Turbo is deprecated but still answers requests until October 23, 2026. Deprecated models get no further updates and are the first to lose capacity.

### How do I migrate from OpenAI GPT-3.5 Turbo to gpt-5.6-terra?

Swap the model id in your API call: replace "gpt-3.5-turbo" with "gpt-5.6-terra". The biggest gotcha is max_tokens: current models use max_completion_tokens and reject the old parameter. Prompts tuned for GPT-3.5 are usually far more verbose than a 5.6-generation model needs.

### Does this apply to gpt-3.5-turbo-0125 too?

Yes. OpenAI GPT-3.5 Turbo is also published as `gpt-3.5-turbo-0125`, `gpt-3.5-turbo-completions`; those ids share the same lifecycle dates as `gpt-3.5-turbo`.

## Sources

- [OpenAI API deprecations](https://developers.openai.com/api/docs/deprecations) — accessed 2026-07-30

---

Canonical page: https://modeldeprecations.dev/openai/gpt-3.5-turbo
JSON: https://modeldeprecations.dev/api/v1/models/openai/gpt-3.5-turbo.json