Llama 4 Scout — Especificaciones
Compiled by Mustafa Ihsan from the vendor’s published documentation · Last updated
| Desarrollador | Meta |
|---|---|
| Tipo | Multimodal (MoE) |
| Modalidad | Texto, imagen → texto |
| Parámetros | 109B totales / 17B activos (MoE) |
| Ventana de contexto | 10M |
| Licencia | Llama 4 Community (restringido en la UE) |
| Pesos abiertos | Sí |
| Lanzado | 2025 |
| Precio de entrada | $0.1 /1M |
| Precio de salida | $0.3 /1M |
| Proveedores de API | Meta, Together, OpenRouter |
Ejecútelo localmente
| VRAM (4 bits) | ~65 GB |
|---|---|
| GPU mínima | H100 de 80 GB / Mac de 128 GB |
What is Llama 4 Scout?
Llama 4 Scout is Meta’s natively multimodal open mixture-of-experts — 109B total
parameters with 17B active across 16 experts, and an industry-leading 10M-token context
window. It fits on a single 80 GB GPU at 4-bit (about 65 GB), or a 128 GB Mac, and ships
under the Llama 4 Community License with the same EU restriction as Maverick.
The 10M context is an order of magnitude beyond the 1M that counts as generous elsewhere,
and it changes what is architecturally possible: entire codebases, full document archives or
long video transcripts can go into a single prompt instead of through a retrieval pipeline.
Whether that is a good idea is a separate question — attention quality across ten million
tokens is not uniform, and retrieval still tends to beat brute force on accuracy and cost —
but for problems where chunking genuinely destroys the signal, Scout is close to unique. That
it does this while fitting on one 80 GB card is the more practical achievement. At $0.10 in /
$0.30 out per million tokens, the API is inexpensive enough to prototype against before
committing to hardware.
Llama 4 Scout pricing: API cost per 1M tokens
| Entrada (por cada millón de tokens) | $0.100 |
|---|---|
| Salida (por cada millón de tokens) | $0.300 |
| Relación salida/entrada | 3× |
| Combinada (4:1 entrada:salida) | $0.140 por 1 millón de tokens |
What Llama 4 Scout costs per month
Gasto mensual real con una mezcla entrada:salida de 4:1, es decir, la proporción que realmente genera una carga de trabajo típica de chat o RAG.
| Carga de trabajo | Tokens/mes | Coste mensual |
|---|---|---|
| Proyecto secundario | 1 millón de tokens de entrada / 0,25 millones de tokens de salida | $0.18 |
| Pequeño equipo | 20 millones de tokens de entrada / 5 millones de tokens de salida | $3.50 |
| Producción | 200 millones de tokens de entrada / 50 millones de tokens de salida | $35 |
Calcule sus propios números en la Calculadora de costos de API de IA.
Cheaper alternatives to Llama 4 Scout
| Modelo | Dólares por millón combinados | Usted ahorra |
|---|---|---|
| Qwen3 32B abierta | $0.120 | 14% cheaper |
| Gemma 3 27B abierta | $0.0960 | 31% cheaper |
¿Autoalojarlo o pagar por la API?
Llama 4 Scout is open-weight, so you can run it yourself. It needs ~65 GB of VRAM at 4-bit (H100 80GB / Mac 128GB). Self-hosting only beats the API once your volume is high enough to keep that hardware busy — the calculadora de autohospedaje frente a API calcula el punto de equilibrio para su volumen de tokens.
Preguntas frecuentes
How much does Llama 4 Scout cost per 1M tokens?
Llama 4 Scout costs $0.100 per 1M input tokens and $0.300 per 1M output tokens. At a typical 4:1 input-to-output mix that blends to about $0.140 per 1M tokens.
How much does Llama 4 Scout cost per month?
A small-team workload of 20M input and 5M output tokens a month costs about $3.50 on Llama 4 Scout. A side project (1M in / 0.25M out) costs roughly $0.18.
What is a cheaper alternative to Llama 4 Scout?
Qwen3 32B is the strongest cheaper option in our database at $0.120 per 1M blended — about 14% less than Llama 4 Scout. It is also open-weight, so self-hosting is an option.
Can I run Llama 4 Scout locally?
Yes. Llama 4 Scout is open-weight and needs about ~65 GB of VRAM at 4-bit quantisation (H100 80GB / Mac 128GB).
Why does Llama 4 Scout charge more for output than input?
Output tokens are generated one at a time and cannot be batched the way a prompt can, so they cost the provider more to serve. Llama 4 Scout charges 3× more for output, which is why prompt-heavy workloads are far cheaper to run than generation-heavy ones.
Los precios corresponden a las tarifas oficiales publicadas para la API principal del modelo y se revisan periódicamente conforme los proveedores los actualicen. No incluyen descuentos por volumen, procesamiento por lotes ni entradas en caché. Compare todos los modelos uno al lado del otro en la Base de datos de modelos de IA o el Clasificación de modelos de lenguaje grande (LLM).