🇺🇸 United States generalistReleased April 2025

Llama 4 Scout

Overview

Llama 4 Scout holds the absolute context record (10 million tokens) on a deployable-on-one-GPU model. Ideal for analyzing entire document bases. Its MoE version activates only 17B params, fast even on accessible hardware.

Skill profile

Arena Elo1322

Verified benchmarks

MMLU80%

HumanEval84%

Three independent measures

What people prefer, what the model can do, what the market uses. Each source has its biases and none is enough on its own: we show them side by side, never blended.

Preference

Which answer do people prefer, blind?

1322Elo

LMArena

Sensitive to answer style and length.

Capability

What can the model do on measured tasks?

6.5/ 100

Source : Artificial Analysis

Aggregates ten evaluations run by Artificial Analysis.

Adoption

What does the market actually use?

Not measured yet

162.2k Hugging Face downloads

Source: OpenRouter (openrouter.ai/rankings). CC BY 4.0

OpenRouter traffic only: measures adoption, not quality, and misses direct use of vendor APIs.

What do these scores mean?

MMLU80%excellent

General knowledge across dozens of academic subjects.

HumanEval84%excellent

Python code generation from specifications.

Strengths

Strong French quality
Very long context
Has a free tier
Open-source and self-hostable

Limitations

API pricing not disclosed

Who is it for

A good fit if…

you want to control cost or self-host
you work in French
your data must stay in the EU

Skip it if…

Ideal use cases

Massive documents (millions of tokens)
Modest self-host
Research
Mid-hardware apps

Access & availability

Free tier availableSelf-hostable (open weights)

Key specifications

Context

10M

Input price

—

Output price

—

Speed

90 tok/s

Price not auditedBenchmarks not audited

What it costs per month

Estimate from sourced API prices and a hardware tier for self-hosting. Assumptions are editable.

M tokens / month % input

Via API

API pricing not published

Self-hosted

642 € / month

amortised hardware 611 € + energy 30.8 € · or cloud GPU 493 €

tier : 80 GB server (A100 / H100) · Q4 ≈ 70 Go

h / day of use

Assumptions: 36-month amortisation, 0.25 €/kWh, 22 days/month, 1 $ = 0.92 €. Excludes labour, software licences, redundancy.

nAIvigate Trust Index

92/100

What the vendor offers and documents for professional use in Europe. An undocumented criterion counts as zero and stays visible: the score rises when the vendor publishes, not when we assume.

19/19 criteria documented · 100 % · 12 of them set by rule

Data100

5/5 · 30 %

Sovereignty77

3/3 · 20 %

Transparency100

4/4 · 15 %

Enterprise100

5/5 · 20 %

Licence80

2/2 · 15 %

See the criteria▾

Data

EU data residencyset by rule

No transfer outside EUset by rule

No training on your dataset by rule

Configurable retentionset by rule

Zero Data Retentionset by rule

Sovereignty

Self-hostable (open weights)

EU hosting offeredset by rule

European vendorcountry: US30

Transparency

Published licence (URL)

Public, sourced pricing

Published compliance commitmentsset by rule

Subprocessor list publishedset by rule

Enterprise

DPA available

SSO / SAMLset by rule

SCIM provisioningset by rule

Audit logsset by rule

SOC 2 / ISO 27001not applicable

Vendor private endpointset by rule

Licence

Commercial use allowed

Licence typellama-4-community60

Full ranking and methodology

Trust Index badge

Show this score in your docs, README or security page. It updates automatically.

[![nAIvigate Trust Index · Llama 4 Scout](https://naivigate.com/badge/llama-4-scout.svg?lang=en)](https://naivigate.com/en/model/llama-4-scout)

Data & compliance

What we know, line by line, with source and date. These lines feed the Data and Enterprise families of the Trust Index above. GDPR compliance depends on your processing, not on the model alone.

EU data residency	Yes
Data processing agreement (DPA)	N/A· self-hosted: data never leaves
Transfer outside the EU	No
Training on your data	No
Configurable retention	N/A· self-hosted: data never leaves
Zero Data Retention	N/A· self-hosted: data never leaves
Self-hostable	Yes
Declared hosting	US/self
Anonymisable data	Yes

Derived from declared hosting, not verified at source.Need an analysis in your context? Shadow AI Sentinel

Quality of this card’s data

74/100

This score measures the quality of the data on this card: price, context, Elo, licence, documented, sourced and fresh. It measures neither the model nor its compliance: for that, see the Trust Index above.

7/8 documented · 3/8 primary sources

API pricingopen weights, no API pricing	missing	—
Context window	nAIvigate assessment	undated
Arena EloLMArena, human preference	independent	6 d ago
Benchmarks (MMLU, GPQA, SWE-bench…)vendor-reported, not cross-checked	nAIvigate assessment	undated
Licence	official	undated
Data & complianceself-hosted: derived	derived	undated
Adoption (Hugging Face)	independent	6 d ago
Hardware requirements (quantizations)computed from model size	derived	6 d ago

Run it locally

Dedicated GPU server162k1k

Quantization	Disk	RAM / VRAM	Typical hardware
Q4 · recommended	62.1 GB	70 GB	96-128GB Mac Studio / 4× GPUs
Q8 · balanced	116.6 GB	130 GB	Server GPU infra (H100/A100…)
FP16 · max quality	218 GB	242 GB	Server GPU infra (H100/A100…)

Estimates for a moderate context. Long contexts need more RAM (KV cache).

🤗 Hugging Face

ollama run llama4:scout

Deploy

Copy-ready commands generated from this card. Adjust context length and GPU count to your hardware.

Estimated memory · FP16 242 Go · Q4 70 Go

Easiest way to try it on a workstation. Install Ollama, then:

ollama run llama4:scout

meta-llama/Llama-4-Scout-17B-16E-Instruct

Advanced data · for experts

Architecture, modalities, detailed cost, full benchmarks

▾

Architecture

MoE

Size

109B (17B actifs MoE)

Cutoff date

Nov 2024

Inputs

text, image

Outputs

text, code

License

llama-4-community

Hosting

US/self

Hallucination score

3/5

Estimated cost (API)

Typical exchange (~3k in / 1k out)

Not disclosed

1M in + 1M out