Models / meta / Llama 4 Maverick 17B 128E Instruct

Llama 4 Maverick 17B 128E Instruct

JSON →
metallm

A 17-billion-parameter mixture-of-experts instruction-tuned model from Meta with 128 experts, optimized for high-performance reasoning and coding.

textimagestreamingcode-generationfunction-callingtool-usejson-modereasoningvision
Specs
Context & limits
131K
context window
8K
max output
Lifecycle
Dates
released2025-04-01
knowledge cutoff2024-12-01
Resources