Llama 4 Maverick

meta-llama/llama-4-maverick

JSON

Specification

Published by the vendor; the platform bills at the target's declared price.

Llama 4 Maverick

open weights tools

Meta meta-llama/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...

Input $/1M: $0.20
Output $/1M: $0.70
Context: 1.0M
Max output: 115K
Released: Apr 5, 2025
Modalities: textimage

Intelligence index-

Coding index16

Agentic index1

On this deployment

No routing target resolves to this model yet. Declare one and the router picks it up without retraining.

Not routable here yet

Add a target with metadata.model set to this id.

targets.yamlyaml

# targets.yaml - declare a target that runs Llama 4 Maverick
targets:
  - id: llama-4-maverick
    kind: llm
    name: Llama 4 Maverick
    metadata:
      model: meta-llama/llama-4-maverick          # links the target to this catalogue entry
    capabilities:
      domains: [general]
      supports_tools: true
      context_window: 1048576
    cost:
      usd_per_1k_tokens: 0.000448

Price among Meta models

Input list price per million tokens; Llama 4 Maverick is the 7th cheapest of 8 listed.

More from Meta

Other models by the same vendor in the catalogue.