MultimodalOpen weights
Llama 4 Scout
Meta
Updated May 21, 2026 · Benchmarks via Artificial Analysis, specs & pricing via OpenRouter · Methodology · Report an error
Specifications
- Type
- Multimodal
- Access
- Open weights
- Released
- April 5, 2025
- License
- Llama 4 Community License
- Parameters
- 109B total / 17B active (MoE)
- Context window
- 10,000,000 tokens
- Knowledge cutoff
- August 31, 2024
- Output speed
- 776 tok/s
- Latency (TTFT)
- 0.31s
- Input
- Text, Image
- Output
- Text
- API pricing
- $0.08 in · $0.30 out / 1M tokens
Capabilities
Function callingStructured output
Benchmarks
Reasoning
| GPQA Diamond | 57.2 |
Coding
| LiveCodeBench | 32.8 |
| MBPP | 67.8 |
General
| MMLU | 79.6 |
| MMLU-Pro | 74.3 |
| Humanity’s Last Exam | 4.3 |
Our take
The smaller Llama 4, notable for an industry-leading 10M-token context window that fits in a single GPU node.