Llama 4 Maverick Instruct
Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total).
text + image → text · made by Meta
$0.2→$0.8per Mtok in / out
DeepInfra
Sold by 3 ways
| Seller | Lane | Rate |
|---|---|---|
| DeepInfra 0 ways | $0.2per Mtok in$0.8per Mtok out | |
| DeepInfra aggregatornot on the seller's list today | standard | $0.2per Mtok in$0.8per Mtok outapi.deepinfra.com · read 2026-08-25 |
| Requesty aggregator | novita | $0.2per Mtok in$0.85per Mtok out$0.2per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| Novita AI 0 ways | $0.27per Mtok in$0.85per Mtok out | |
| Novita AI aggregatornot on the seller's list today | standard | $0.27per Mtok in$0.85per Mtok outapi.novita.ai · read 2026-08-25 |
About
Llama 4 Maverick Instruct — a text model from Meta.
It takes text and images and returns text, with a context window of 1,048,576 tokens. It was published in April 2025. The catalogue files it under chat.
Every current figure
- Maker
- Meta
- Register
- model
- Takes
- text + image
- Returns
- text
- Context
- 1,048,576 tokens
- Longest answer
- 8,192 tokens
- Published
- April 2025
- Licence
- not read
Known as 1 name
novita/meta-llama/llama-4-maverick-17b-128e-instruct-fp8