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.
Índice de Qualidade
18.4
214th de 444
Top 48%
Índice de Código
15.6
188th de 354
Top 53%
Índice de Matemática
19.3
210th de 268
Top 78%
Preço/1M
$0.49
391st mais barato
62% acima da mediana
Top 57%
Velocidade
127 tok/s
Top 20%
TTFT
0.48s
Janela de Contexto
1.0M
8th maior
Top 6%
Entrada
$0.31
por 1M tokens
Saída
$0.91
por 1M tokens
Combinado
$0.49
por 1M tokens
Mais barato que 43% dos modelos. Preço mediano é $0.30/1M tokens.
Diário
$0.49
Mensal
$14.61
127
tokens/seg
Mais rápido que 80% dos modelos
0.48
segundos
Mais rápido que 47% dos modelos
0.48
segundos
Mais rápido que 49% dos modelos
Mediana do Mercado
45 tok/s
180% mais rápido
TTFT Mediano
0.42s
14% mais lento
Vazão/Dólar
261
tok/s por $/1M
Comparação de Velocidade
Janela de Contexto
1.0M
tokens
Maior que 94% dos modelos
Saída Máxima
16K
tokens
2% do contexto