# meta-llama/llama-4-maverick-17b-128e-instruct

Calls to the model meta-llama/llama-4-maverick-17b-128e-instruct on Groq stopped working on March 9, 2026.

## The error Groq returns once meta-llama/llama-4-maverick-17b-128e-instruct is gone

HTTP 404

```json
{"error":{"message":"The model `meta-llama/llama-4-maverick-17b-128e-instruct` does not exist or you do not have access to it.","type":"invalid_request_error","code":"model_not_found"}}
```

## What to use instead

Update your requests to use the recommended replacement model openai/gpt-oss-120b.

- API: Groq (api.groq.com)
- Model: meta-llama/llama-4-maverick-17b-128e-instruct
- Kind: Model retirement
- Takes effect: March 9, 2026
- Calls after that day: fail
- Calls it hits: `model meta-llama/llama-4-maverick-17b-128e-instruct`
- Source: [Deprecations page](https://console.groq.com/docs/deprecations)
- Verified: November 6, 2026, against Groq's own pages
- Page: https://www.okapi.bot/models/groq/meta-llama/llama-4-maverick-17b-128e-instruct/

## Groq's words

> Deprecated Model: meta-llama/llama-4-maverick-17b-128e-instruct; Shutdown Date: 03/09/26; Recommended Replacement Model ID: openai/gpt-oss-120b

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