Back

Meta: Llama 4 Maverick

Llama4
Input: text
Input: image
Output: text
Released: Apr 5, 2025Updated: May 16, 2025

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). It supports multilingual text and image input, and produces multilingual text and code output across 12 supported languages. Optimized for vision-language tasks, Maverick is instruction-tuned for assistant-like behavior, image reasoning, and general-purpose multimodal interaction.

Maverick features early fusion for native multimodality and a 1 million token context window. It was trained on a curated mixture of public, licensed, and Meta-platform data, covering ~22 trillion tokens, with a knowledge cutoff in August 2024. Released on April 5, 2025 under the Llama 4 Community License, Maverick is suited for research and commercial applications requiring advanced multimodal understanding and high model throughput.

1,048,576 Token Context

Process and analyze large documents and conversations.

Advanced Coding

Improved capabilities in front-end development and full-stack updates.

Agentic Workflows

Autonomously navigate multi-step processes with improved reliability.

Vision Capabilities

Process and understand images alongside text inputs.

Available On

ProviderModel IDContextMax OutputInput CostOutput CostThroughputLatency
DeepInfradeepInfra1049K16K$0.15/M$0.60/M95.7 t/s278 ms
Parasailparasail1049K1049K$0.15/M$0.85/M160.1 t/s255 ms
KlusterklusterAi1049K1049K$0.16/M$0.80/M120.2 t/s767 ms
NovitanovitaAi1049K1049K$0.17/M$0.85/M65.4 t/s424 ms
Lambdalambda1049K1049K$0.18/M$0.60/M153.1 t/s287 ms
BaseTenbaseten1000K131K$0.19/M$0.72/M203.6 t/s153 ms
Cent-MLcentMl1049K1049K$0.20/M$0.20/M72.3 t/s257 ms
Groqgroq131K8K$0.20/M$0.60/M1193.5 t/s254 ms
NCompassnCompass400K400K$0.20/M$0.70/M142.9 t/s94 ms
Fireworksfireworks1049K-$0.22/M$0.88/M95.8 t/s489 ms
GMICloudgmiCloud1049K-$0.25/M$0.80/M155.3 t/s529 ms
Togethertogether1049K-$0.27/M$0.85/M88.0 t/s299 ms
Googlevertex524K-$0.35/M$1.15/M102.4 t/s795 ms
DeepInfradeepInfraTurbo8K-$0.50/M$0.50/M411.8 t/s475 ms
SambaNovasambaNova131K4K$0.63/M$1.80/M640.9 t/s952 ms
Standard Pricing
Input Tokens
$0.00000015

per 1K tokens

Output Tokens
$0.0000006

per 1K tokens

Image Processing
$0.0006684

per image

Do Work. With AI.