Scout
About this model
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 (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens. Built for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025.
Performance Tier
Scout is a balanced model from Llama 4 : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plans| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.080 |
| Output (completion) | $0.300 |
Capabilities
Benchmarks
Recommended Use Cases
Strengths
- Massive 10M token context window — largest in the industry
- Efficient MoE architecture (17B active / 109B total)
- Strong instruction following (IFEval 85.2%)
- Open-weight with commercially permissive license
Limitations
- Less capable than Maverick on complex reasoning
- Quality may degrade at extreme context lengths
Resources
This model may use your data for training
