Inference
EN
Meta

Llama 4 Scout

Llama 4 Scout is a standard model from Meta. 14 benchmarks count toward its index, in 7 categories.

  • Open weights
  • Standard
  • Input: text · image
  • Released 2025-04
RankedEnough results for an index.
IndexAll categories, out of 100.26.8±4.8Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.63/sMedian 67/s
First tokenSeconds to the first output token.0.4 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.10Median $0.60
Output / 1MUS dollars per 1M output tokens.$0.30Median $2.55
ContextMaximum tokens in one request.1.31MMedian 262K
EloLMArena rating and rank.—Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
meta-llama/llama-4-scout
Max outputMaximum output tokens per request.
16K
Knowledge cutoffLast date of training data.
31 Aug 2024
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
—
WeightsLink to the published weights.
Hugging Face (opens in a new tab) ↗
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
7 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
Deep InfraHost$0.10$0.30328K
OpenRouterRouter$0.10$0.30328K
Novita AIHost$0.18$0.59131K
Amazon BedrockDeprecatedCloud$0.17$0.6610M
Google Vertex AICloud$0.25$0.701.31M
Oracle CloudCloud$0.72$0.72192K
Cloudflare Workers AICloud$0.27$0.85131K

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Run locally

108.64B parameters
16-bit
229.1 GB
8-bit
122.2 GB
4-bit
70.2 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy fits in 128 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted23.6217 / 218
CodingCode writing and repair.1 counted23.6—
ReasoningLogic problems and puzzles.4 counted32.4201 / 227
MultimodalTasks with images and text.1 counted24.0—
KnowledgeFacts and expert knowledge.2 counted28.0245 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted25.9—
MathMath problems.3 counted24.2153 / 168

Results

14 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH level 5Math62.3% ±1.241 / 88
GPQA diamond · Epoch AIKnowledge51.8% ±3.2141 / 217
Critical Physics TasksReasoning0.0%150 / 214
Artificial Analysis IFBenchInstruction39.5%106 / 146
GDPval-AA normalizedAgentic0.0%101 / 133
OTIS Mock AIME 2024-2025Math7.8% ±3.0160 / 200
Artificial Analysis GPQA DiamondKnowledge58.7%160 / 199
Artificial Analysis Long Context ReasoningReasoning27.7%184 / 217
Artificial Analysis MMMU-ProMultimodal52.9%101 / 113
Artificial Analysis Humanity's Last ExamKnowledge3.8%202 / 222
ARC-AGI-2 (semi-private)Reasoning0.0%120 / 131
τ²-Bench Tool-Agent-User EvaluationAgentic15.5%139 / 149
FrontierMath-2025-02-28-PrivateMath0.0%68 / 69
Artificial Analysis SciCodeCoding21.3%113 / 114
ARC-AGI-1 (semi-private)Reasoning0.5%125 / 126
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
ARC-AGI-2 (public eval)Reasoning0.0%102 / 122
ARC-AGI-1 (public eval)Reasoning2.4%115 / 116

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Index27.5 → 26.8
23 Oct 202612 Nov 2026
Output / 1M$0.30 → $0.30
23 Oct 202612 Nov 2026

Questions

How much does Llama 4 Scout cost?
$0.10 per 1M input tokens, $0.30 per 1M output tokens.
How good is Llama 4 Scout?
Its index is 26.8 out of 100. That is place 264 of 283 ranked models.
What is Llama 4 Scout best at?
Reasoning: 32.4 out of 100, place 201 of 227.
How fast is Llama 4 Scout?
63 output tokens per second, 0.4 s to the first token.
What is the context window of Llama 4 Scout?
1.31M tokens.
Is Llama 4 Scout open source?
Yes. Its weights are published.
When was Llama 4 Scout released?
April 2025.

Badge

Llama 4 Scout on Inference360: place 264 of 283, index 26.8
HTML
<a href="https://inference360.ai/models/llama-4-scout"><img src="https://inference360.ai/badges/llama-4-scout.svg" alt="Llama 4 Scout on Inference360: place 264 of 283, index 26.8" height="20"></a>
Markdown
[![Llama 4 Scout on Inference360: place 264 of 283, index 26.8](https://inference360.ai/badges/llama-4-scout.svg)](https://inference360.ai/models/llama-4-scout)

Compare with

Llama 3.1 405B27.0Phi-4-reasoning27.1GPT-4.1 nano26.3GPT-4o mini26.2Llama 4 Maverick34.4

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from Meta

All models
Llama 4 Maverick34.4Llama Guard 4 12BUnrankedLlama 3.3 70B InstructUnrankedLlama 3.3 70B28.2Llama 3.2 1B InstructUnranked