Back to the index
Meta AI · Llama
No. 19

Llama 4 Scout

Ultra-long 10M-token context, efficient single-GPU open model.

OpenOpen weightsText, Image
Field notes
Context window
10Mtokens
Max output
16Ktokens
Input price
$0.10/ 1M tokens
Output price
$0.30/ 1M tokens
Modalities
Text, Image
Knowledge cutoff
Aug 2024
Released
Apr 2025
Parameters
109B
License
Llama 4 Community License
Approx. speed
Overview

What this model is

Llama 4 Scout is the smaller, faster member of the Llama 4 herd (109B-total / 17B-active MoE) that fits on a single high-end GPU and ships a headline 10M-token context window. It is ideal for cheap long-document and retrieval workloads, though effective reasoning quality over the full window is far below the nominal length. A pragmatic open model for scale, not for hard agentic work.

Strengths

  • Industry-leading 10M-token context window
  • Runs on a single GPU / very cheap
  • Open weights and fine-tunable
  • Fast throughput for its class

Trade-offs

  • Reasoning/coding well below the frontier
  • Effective recall degrades long before 10M tokens
  • Weak on autonomous agentic tasks
  • Vision is basic

Best for

  • Long-document summarization and RAG
  • High-throughput cheap inference
  • On-prem/self-hosted deployments
  • Fine-tuning experiments

Not ideal for

  • Agentic coding
  • Complex multi-step reasoning
Capabilities

Capability profile

Normalized 0–100 scores, comparable across the whole catalog.

Reasoning58Coding52Math56Writing66Knowledge70Speed88Agentic50Vision64Multilingual76Long Context82
Reasoning
58
Coding
52
Math
56
Writing
66
Knowledge
70
Speed
88
Agentic
50
Vision
64
Multilingual
76
Long Context
82
Benchmarks

How it scores

Public benchmark results, with independent third-party results where available. Bars normalize percentages to 100 and Elo ratings to a 1500 ceiling.

Independent index
Artificial Analysis Intelligence Index
Composite of ~9–10 independent evals · Artificial Analysis
10/ 100
BenchmarkResult
Artificial Analysis Intelligence Indexindependent
GeneralArtificial Analysis
10%
τ-benchindependent
Agenticτ-bench
62.3%
MMLU-Pro
KnowledgeVendor-reported
74.3%
GPQA Diamondindependent
ReasoningIndependent aggregators
57.2%
MMMU
VisionVendor-reported
69.4%
Alternatives

Comparable models

Models in a roughly similar class — worth weighing against this one.