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 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
Provider | Model ID | Context | Max Output | Input Cost | Output Cost | Throughput | Latency |
---|---|---|---|---|---|---|---|
DeepInfra | deepInfra | 1049K | 16K | $0.15/M | $0.60/M | 95.7 t/s | 278 ms |
Parasail | parasail | 1049K | 1049K | $0.15/M | $0.85/M | 160.1 t/s | 255 ms |
Kluster | klusterAi | 1049K | 1049K | $0.16/M | $0.80/M | 120.2 t/s | 767 ms |
Novita | novitaAi | 1049K | 1049K | $0.17/M | $0.85/M | 65.4 t/s | 424 ms |
Lambda | lambda | 1049K | 1049K | $0.18/M | $0.60/M | 153.1 t/s | 287 ms |
BaseTen | baseten | 1000K | 131K | $0.19/M | $0.72/M | 203.6 t/s | 153 ms |
Cent-ML | centMl | 1049K | 1049K | $0.20/M | $0.20/M | 72.3 t/s | 257 ms |
Groq | groq | 131K | 8K | $0.20/M | $0.60/M | 1193.5 t/s | 254 ms |
NCompass | nCompass | 400K | 400K | $0.20/M | $0.70/M | 142.9 t/s | 94 ms |
Fireworks | fireworks | 1049K | - | $0.22/M | $0.88/M | 95.8 t/s | 489 ms |
GMICloud | gmiCloud | 1049K | - | $0.25/M | $0.80/M | 155.3 t/s | 529 ms |
Together | together | 1049K | - | $0.27/M | $0.85/M | 88.0 t/s | 299 ms |
vertex | 524K | - | $0.35/M | $1.15/M | 102.4 t/s | 795 ms | |
DeepInfra | deepInfraTurbo | 8K | - | $0.50/M | $0.50/M | 411.8 t/s | 475 ms |
SambaNova | sambaNova | 131K | 4K | $0.63/M | $1.80/M | 640.9 t/s | 952 ms |
per 1K tokens
per 1K tokens
per image