Model dossier

Meta: Llama 4 Maverick

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...

Context1,048,576
Max output16,384
Pricing viewTVP price

Meta: Llama 4 Maverick

Model ID
meta-llama/llama-4-maverick
Canonical slug
meta-llama/llama-4-maverick-17b-128e-instruct
Knowledge cutoff
2024-08-31
Created
4/5/2025
textimagetoolsstructured
text+image->texttextimage
Context1,048,576
Max output16,384
TokenizerLlama4