Model Information

Display Name: Llama 4 Scout

API Model ID: meta-llama/llama-4-scout

Category: Image To Text

Description: Llama 4 Scout is Meta's natively multimodal Mixture-of-Experts model — 17B active parameters across 16 experts (109B total) — built for long-context text and image understanding at low cost. **Key Features:** - 327,680-token context window - Up to 16,384 output tokens - Natively multimodal: text and image input - Mixture-of-Experts: 17B activated / 109B total parameters - Function/tool calling and structured outputs (JSON schema) - Non-reasoning: direct answers, no thinking budget **Best For:** - Long-document and long-conversation understanding - Image understanding, OCR and visual Q&A - High-volume tool-using workflows where cost per token matters - General-purpose chat at a low price point

Context Window: 327,680 tokens

Max Output: 16,384 tokens

How to Use This Model

To use Llama 4 Scout via the HInow.ai API, use the model ID: meta-llama/llama-4-scout

API Request Example (Chat/Text)

POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "meta-llama/llama-4-scout",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}
              

API Request Example (Image Generation)

POST https://api.hinow.ai/v1/images
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "meta-llama/llama-4-scout",
  "prompt": "Your image description here"
}
              

Pricing

• input: $0.35
• output: $1.05

Available Parameters

• temperature: Controls randomness (0-2). Default: 0.7 (Options: 0, 0.3, 0.5, 0.7, 1.0)
• top_p: Nucleus sampling (0-1). Default: 0.95 (Options: 0.1, 0.5, 0.8, 0.9, 0.95, 1.0)
• top_k: Top-k sampling. 0 disables (Options: 0, 10, 40, 100)
• min_p: Minimum probability threshold relative to the most likely token (Options: 0, 0.05, 0.1)
• max_tokens: Max tokens to generate (1-16384). Default: 8192 (Options: 512, 1024, 2048, 4096, 8192, 16384)
• frequency_penalty: Penalizes tokens by frequency (-2 to 2). Default: 0 (Options: 0, 0.5, 1.0)
• presence_penalty: Penalizes tokens already present (-2 to 2). Default: 0 (Options: 0, 0.5, 1.0)
• repetition_penalty: Penalizes repetition (>1 reduces it). Default: 1.0 (Options: 1.0, 1.05, 1.1)
• seed: Integer seed for reproducible sampling
• tools: Function/tool definitions for agentic workflows
• response_format: Output format (structured outputs supported) (Options: text, json_object, json_schema)
• stop: Sequences where the API stops generating further tokens

Quick Reference

To use this model, set: "model": "meta-llama/llama-4-scout"

Featured: No

Documentation: https://hinow.ai/models/meta-llama/llama-4-scout

API Endpoint: https://api.hinow.ai/v1