When was Llama 4 Scout released?

Llama 4 Scout was released on April 5, 2025 by Meta. This is the official Llama 4 Scout release date tracked on LLM Stats.

How much does Llama 4 Scout cost?

Llama 4 Scout pricing starts at $0.08 per million input tokens and $0.30 per million output tokens via DeepInfra, the lowest price among tracked providers.

Is Llama 4 Scout available via API?

Yes, Llama 4 Scout is available via API. See the official documentation for authentication and endpoint details. It is served by 6 providers tracked on LLM Stats.

How big is Llama 4 Scout?

Llama 4 Scout has 109 billion parameters. It was trained on 40.0 trillion tokens. It ships as an open-weight model, so you can download and run it on your own hardware.

Who created Llama 4 Scout?

Llama 4 Scout was created by Meta.

What is the license for Llama 4 Scout?

Llama 4 Scout is released under the Llama 4 Community License Agreement license. This is an open-source / open-weight license that permits self-hosting.

Is Llama 4 Scout multimodal?

Yes, Llama 4 Scout is multimodal and can accept both text and images as input.

What is Llama 4 Scout latency?

Llama 4 Scout p95 time to first token is 0.31 seconds via DeepInfra over the trailing 7 days. Lower time to first token means the model begins responding sooner for chat, agents and API workloads.

Where can I use Llama 4 Scout?

Llama 4 Scout is available through 6 providers including DeepInfra, Lambda, Novita, and 3 more.

What models should I compare Llama 4 Scout against?

Common Llama 4 Scout comparisons include Llama 4 Scout vs Grok-2, Llama 4 Scout vs Gemini 1.5 Pro, Llama 4 Scout vs Phi 4 Reasoning. Compare them side by side for benchmark scores, pricing, context window, latency and API availability.

Llama 4 Scout API Pricing, Context Window & Benchmarks

Name: Llama 4 Scout
Author: Meta

Llama 4 Scout: API Pricing, Context Window & Benchmarks

Llama 4 Scout is a language model from Meta, released in April 2025, with multimodal input.

Llama 4 Scout is a natively multimodal model capable of processing both text and images. It features a 17 billion activated parameter (109B total) mixture-of-experts (MoE) architecture with 16 experts, supporting a wide range of multimodal

Input

TextImage

Output

Text

Llama 4 Scout benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

High-confidence performance for Llama 4 Scout across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.

Performance by conversation depth

How Llama 4 Scout holds up as conversations get longer.

Quality Tracker

Llama 4 Scout Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Mon Aug 31 2026

Notice missing or incorrect data?

Llama 4 Scout pricing

Providers

Llama 4 Scout starts at $0.0800 per million input tokens and $0.300 per million output tokens via DeepInfra. See all 6 providers below with their per-token pricing, latency, throughput, and modality support.

Provider	Input $/M	Cached input $/M	Output $/M	Context in / out	TTFT p95 s	Output p5 c/s	Modalities in / out
DeepInfra	$0.0800	—	$0.300	10.0M/10.0M	0.31	—	/
Lambda	$0.0800	—	$0.300	10.0M/10.0M	0.43	—	/
Novita	$0.100	—	$0.500	10.0M/10.0M	0.85	—	/
Groq	$0.110	—	$0.340	10.0M/10.0M	1.08	—	/
Fireworks	$0.150	—	$0.600	10.0M/10.0M	0.53	—	/
Together	$0.180	—	$0.590	10.0M/10.0M	0.54	—	/

Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.

Loading chart...

Llama 4 Scout model size

Llama 4 Scout has 109 billion parameters and was trained on 40 trillion tokens. See how it compares to other models in the same parameter range.

ParametersTraining tokens

109B

40Ttokens

367× tokens-to-params ratio

Very large (80–200B)

109B

1B7B70B405B

Llama 4 Scout context window

Input and output token limits for Llama 4 Scout, plus how it ranks on long-context understanding.

InputOutput

10Mtokens

≈ 15.0k pages of text

10M

8K128K1M

Llama 4 Scout API

Available from the model provider

Llama 4 Scout has an official provider API. It is not currently routed through the LLM Stats gateway.

Read the official API documentation

Llama 4 Scout latency

Llama 4 Scout time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.

Llama 4 Scout examples

Recent arena outputs from Llama 4 Scout, picked from the highest-ranked matchups.

Llama 4 Scout license

Llama 4 Scout is released under the Llama 4 Community License Agreement license, which restricts commercial use, has 109.0B parameters.

License: Llama 4 Community License Agreement; Non-commercial
Parameters: 109.0B

Llama 4 Scout resources

Official sources for Llama 4 Scout: api documentation, official playground, source repository.

Llama 4 Scout vs other models

The most-compared alternatives to Llama 4 Scout are Grok-2, Gemini 1.5 Pro, Phi 4 Reasoning. Open any pair side-by-side for benchmarks, pricing, context, and latency.