Llama 4 Scout

Apr 2025
Llama 4 Scout
Compare in
on
$0.11per task?
$0.10 / $0.30per 1M in / out?
82 tok/sfirst token in 0.8 s?
10Mcontext

Rankingin non-thinking mode?

Full ranking →
1Gemini 4 Argon77.52MiniMax M3.1 Flash72.53Sonnet 5.571.84Muse Wonder Full71.35Opus 5.569.5
···
130GPT OSS 120B0.2131Llama 4 Scout0.2131GPT OSS 20B0.2

Strong at

C
Coding#75 of 132Agents & work tasks#122 of 175Tool calling#207 of 242Reasoning#315 of 334Science#229 of 312Long context#222 of 273Knowledge#220 of 318Instructions#160 of 245Vision#106 of 124

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperMiMo v2.6 FlashthinkingOpen×335.1-44%$0.062Context: 10M → 1M+5BetterGPT-6 Lunalow×63-96%$0.005Context: 10M → 1.05M+4OpenQwen3.8 Flashthinking×292.5+232%$0.37Context: 10M → 262K+7CheaperGPT-5.6 Lunaxhigh×222.9-24%$0.085Context: 10M → 1.05MCheaperClaude Haiku 5.5high×176.5-29%$0.079Context: 10M → 1M+6

Run it yourself

estimate
Fits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✓ 80 GB GPU · compressed
LicenceLlama 4 licencecommercial use with conditions

Benchmarks

in Agents & Tool Use
Terminal-Bench 2.13.7#140
Terminal-Bench 4.00#75
Terminal-Bench Hard (version unstated)1.5#197
GDPval-AA (work tasks) 2.10#122
AA-LCR (long context) 1.127.7#224
τ²-bench · AA15.5#210
τ³-Banking · AA3.3#131
AutomationBench-AA0.2#131
AA-Briefcase 1.1 · Elo0#129

FAQ

How much does Llama 4 Scout cost?›

$0.10 per 1M input tokens and $0.30 per 1M output tokens (cheapest host: DeepInfra). A typical benchmark task costs about $0.11 at default effort.

Is Llama 4 Scout free to use?›

There is no free API; the cheapest paid price is $0.10 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Llama 4 Scout, and what is it best at?›

#315 of 334 in Chat & Knowledge, #75 of 132 in Coding, #131 of 133 in Agents & Tool Use, #138 of 195 in Writing & Language, #106 of 124 in Vision & Documents.

Is Llama 4 Scout better than GPT OSS 120B?›

On AutomationBench-AA, Llama 4 Scout scores 0.2 vs 0.2 — about the same. It is 22% more expensive per task.

How can I use Llama 4 Scout?›

Through the API of DeepInfra, Google, Novita.

What is the context window of Llama 4 Scout?›

10M tokens, with up to 16K output tokens.

When was Llama 4 Scout released? Is there a newer version?›

Released 2025-04-05. It is the current version in its line.

Is Llama 4 Scout open source?›

Yes, the weights are open (llama4 licence).

Can I run Llama 4 Scout locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can Llama 4 Scout read images?›

It accepts text and images as input.

What is Llama 4 Scout?›

Llama 4 Scout is an open-weights language model from Meta, released Apr 2025.

What are the best alternatives to Llama 4 Scout?›

MiMo v2.6 Flash (cheaper), GPT-6 Luna (better), Qwen3.8 Flash (open weights).