Llama 4 Maverick, not served here yet
Llama 4 Maverick is Meta's natively multimodal mixture-of-experts: 17B active parameters across 128 experts, 400B in total, text and images in. OpenRelay does not serve it yet; tell us if you want it.
- Text and image in, text out
- Not served yet
- No public rate
Not served yet.
GET /v1/models lists Llama 4 Maverick with status request_access, but nothing serves it: there is no backend behind the id and no public rate. A request tells us someone wants it.
openrelay/llama-4-maverickTell us you want Llama 4 Maverick
Leave a work email. We email you if we start serving it, and nothing else goes to this list.
The spec sheet.
What Meta has published about the model, with the source on each row.
- On OpenRelay
- openrelay/llama-4-maverick: Text and image in, text out. Listed with status request_access, not served. GET /v1/models
- Parameters
- 400B total, 17B activated, 128 experts Meta model card
- Context
- 1M tokens Meta model card
- Modality
- Multilingual text and image in, multilingual text and code out Meta model card
- Languages
- Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, Vietnamese Meta model card
- License
- Llama 4 Community License Agreement Meta model card
- Released
- April 5, 2025 Meta model card
Llama 4 Maverick, answered.
Can I call Llama 4 Maverick through OpenRelay?
Not today. openrelay/llama-4-maverick appears on GET /v1/models with status request_access, but nothing serves it: a chat completions call returns 403 model_access_required. The form on this page tells us you want it.
What does Maverick take in?
Text and images, per Meta's model card, with text and code out. The OpenRelay catalog lists it the same way: text and image in, text out.