Loading...
Loading...
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.
Quality Index
18.4
213th of 442
Top 48%
Coding Index
15.6
187th of 352
Top 53%
Math Index
19.3
210th of 268
Top 78%
Price/1M
$0.47
384th cheapest
53% above median
Top 57%
Speed
130 tok/s
Top 19%
TTFT
0.47s
Context Window
1.0M
8th largest
Top 6%
Input
$0.27
per 1M tokens
Output
$0.85
per 1M tokens
Blended
$0.47
per 1M tokens
Cheaper than 43% of models. Median price is $0.31/1M tokens.
Daily
$0.47
Monthly
$14.25
130
tokens/sec
Faster than 81% of models
0.47
seconds
Faster than 46% of models
0.47
seconds
Faster than 49% of models
Market Median
46 tok/s
186% faster
Median TTFT
0.42s
12% slower
Throughput/Dollar
274
tok/s per $/1M
Speed Comparison
Context Window
1.0M
tokens
Larger than 94% of models
Max Output
16K
tokens
2% of context