Meta
Llama 4 Scout
Open multimodal Llama model for long-context analysis and efficient agents
Model facts
- Context window
- 1.31M tokens
- Maximum output
- 16K tokens
- Cheapest paid input
- $0.1 per 1M tokens
- Cheapest paid output
- $0.3 per 1M tokens
- Open weights
- yes
- Providers
- 4
Capabilities and modalities
tool calling, structured output, attachments, temperature control, open weights, text, image.
Local hardware estimate
At 4K context: Q4_K_M 67.5 GiB, Q8_0 118.1 GiB, F16 212.9 GiB working memory.
- Parameters
- 108.6 billion
- Architecture
- llama4
- Model license
- other
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| default | 10.3 | 8.2 | 14.0 | 98.5 |
Observed price history
- 2026-08-24: $0.1 input / $0.3 output per 1M tokens
- 2026-08-26: $0.1 input / $0.3 output per 1M tokens
- 2026-09-01: $0.1 input / $0.3 output per 1M tokens
API providers
- OpenRouter — model id meta-llama/llama-4-scout; $0.1 input / $0.3 output per 1M tokens; provider documentation
- Kilo Gateway (Meta) — model id meta-llama/llama-4-scout; $0.1 input / $0.3 output per 1M tokens; provider documentation
- NanoGPT — model id meta-llama/llama-4-scout; $0.085 input / $0.46 output per 1M tokens; provider documentation
- NovitaAI — model id meta-llama/llama-4-scout-17b-16e-instruct; $0.18 input / $0.59 output per 1M tokens; provider documentation