Meta: Llama 4 Scout: AI model
Llama 4 Scout is Meta's efficient multimodal Mixture-of-Experts model with 17B active parameters and a very large context window. Reachable from inside Iran through Hormouz.
- Context window:
- 1,310,720 tokens
- Input:
- $0.18/1M tokens
- Output:
- $0.59/1M tokens
- Type:
- text, image
- Vision
- Tools
- Streaming
- Long context
Llama 4 Scout is a Mixture-of-Experts (MoE) language model from Meta that activates 17 billion of a total 109 billion parameters. It accepts native multimodal input (text and image) and produces text and code output across 12 languages.
Scout uses 16 experts per forward pass with a focus on high efficiency, and is designed for assistant-style interaction and visual reasoning. It is built with early fusion for seamless multimodality and is served through Hormouz behind an OpenAI-compatible API.
Pricing is usage-based; the highest rate across providers is $0.25 per million input tokens and $0.70 per million output tokens. Hormouz provides stable access to the model from inside Iran.
Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...
Quickstart
curl https://api.hormouz.net/v1/chat/completions \
-H "Authorization: Bearer ng-YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta-llama/llama-4-scout",
"messages": [{"role": "user", "content": "Hello!"}]
}'Specs
| Vendor | Meta |
|---|---|
| Type | Text, Image |
| Context window | 1,310,720 tokens |
| Pricing | $0.18/M in · $0.59/M out |
| API | Chat Completions |
| Capabilities | Streaming · Vision · Tools |
Benchmarks
Reasoning
Coding
Agentic
Source: Artificial Analysis
Design Arena
How Design Arena works ↗Metrics sourced from Design Arena
Code example
curl https://api.hormouz.net/v1/chat/completions \
-H "Authorization: Bearer ng-YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta-llama/llama-4-scout",
"messages": [{"role": "user", "content": "Hello!"}]
}'from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
model="meta-llama/llama-4-scout",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
model: "meta-llama/llama-4-scout",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);Buy the Meta: Llama 4 Scout API, paid in rial
- Buy credit with an Iranian bank card; no foreign card or account needed.
- One Hormouz key reaches Meta: Llama 4 Scout and 600+ other models, and you pay only for what you use.
- No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Use via Hormouz
Connect to this model and 600+ others with one API key, then see usage down to each request.
Request company accessFAQ
- Does Llama 4 Scout understand images?
- Yes. Scout is a multimodal model and can process text and image input together, which fits visual reasoning and assistant-style interaction.
- How does Llama 4 Scout differ from Llama 4 Maverick?
- Scout uses 16 experts with a focus on high efficiency and is more cost-effective, while Maverick uses 128 experts for the highest capacity on heavier multimodal work.
- How much does Llama 4 Scout cost?
- Billing is usage-based; the highest rate across providers is $0.25 per million input tokens and $0.70 per million output tokens.
- How do I reach Llama 4 Scout from inside Iran?
- Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card or circumvention tooling required.
- Is Llama 4 Scout compatible with OpenAI-style code?
- Yes. Hormouz exposes an OpenAI-compatible API. Point your base URL at Hormouz and swap in your Hormouz key. Your existing SDKs and code run without a rewrite.
Updated 2026-10-13