#### Migration Guides

# Grok Model Deprecation on May 15, 2026

As we continue advancing Grok, we are retiring several earlier models to focus fully on our newest generation. **Effective May 15, 2026 at 12:00pm PT**, the following models will be deprecated from the xAI API:

* `grok-4-1-fast-reasoning`
* `grok-4-1-fast-non-reasoning`
* `grok-4-fast-reasoning`
* `grok-4-fast-non-reasoning`
* `grok-4-0709`
* `grok-code-fast-1`
* `grok-3`
* `grok-imagine-image-pro`

After **May 15, 2026 at 12:00pm PT**, requests to the models listed above will no longer work.

We know transitions like this can be disruptive, and we truly appreciate your understanding as we make this shift. Our goal is to deliver the best possible performance and capabilities to everyone building with us, and we have some exciting model releases planned in the coming weeks!

## Recommended Replacements

We have newer, more capable models ready for every workload. In most cases, migrating is as simple as changing the `"model"` field in your API request.

| Deprecated Model | Recommended Replacement |
|---|---|
| `grok-4-1-fast-reasoning` | `grok-4.3` |
| `grok-4-1-fast-non-reasoning` | `grok-4.20-non-reasoning` |
| `grok-4-fast-reasoning` | `grok-4.3` |
| `grok-4-fast-non-reasoning` | `grok-4.20-non-reasoning` |
| `grok-4-0709` | `grok-4.3` |
| `grok-code-fast-1` | `grok-4.3` |
| `grok-3` | `grok-4.3` |
| `grok-imagine-image-pro` | `grok-imagine-image` |

### Reasoning workloads

If you are using `grok-4-1-fast-reasoning`, `grok-4-fast-reasoning`, or `grok-4-0709`, we recommend migrating to **`grok-4.3`**.

Grok 4.3 is the fastest, most intelligent model we have ever built. It tops the leaderboards in agentic tool calling and instruction following, and includes:

* 1 million token context window
* 3 reasoning effort levels (low, medium, and high)
* Priced at $1.25 / 1M input and $2.50 / 1M output

### Non-reasoning workloads

If you are using `grok-4-1-fast-non-reasoning` or `grok-4-fast-non-reasoning`, we recommend migrating to **`grok-4.20-non-reasoning`**.

For less latency-sensitive workloads, consider using `grok-4.3` with [reasoning\_effort](/developers/model-capabilities/text/reasoning#the-reasoning_effort-parameter) set to `low`.

### Code workloads

If you are using `grok-code-fast-1`, we recommend migrating to **`grok-4.3`**, which delivers significantly improved agentic coding and web dev capabilities.

## Need Help?

We are here to support you through this migration. If you have any questions or need assistance, please reach out to [support@x.ai](mailto:support@x.ai).

For general information on model lifecycle and deprecation policies, see [Migrating to New Models](/developers/migration/models).