About Llama 4 Scout
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...
Specifications
- Provider
- Meta
- Context Length
- 10,000,000 tokens
- Input Types
- text, image
- Output Types
- text
- Category
- meta-llama/llama-4-scout-17b-16e-instruct
- Added
- 4/5/2025
Benchmark Performance
How Llama 4 Scout compares to its closest rivals across industry benchmarks
Evaluates AI ability to resolve real GitHub issues from Python repositories
Metric: % Resolved
#21
gpt-oss-120b
26.0
#25
Gemini 2.0 flash
13.5
#27
Qwen2.5-Coder 32B Instruct
9.0
Comprehensive comparison of AI models across intelligence, price, speed, and latency
Metric: Intelligence Index
#120
Solar Pro 2
30.0
#121
Qwen3 Omni 30B A3B
30.0
#123
Mistral Small 3.2
29.0
#124
Ministral 8B (Dec '25)
28.0
#126
Llama 3.1 405B
28.0
#127
Llama 3.3 70B
28.0
#128
Devstral Medium
28.0
#129
Ling-mini-2.0
28.0
Frequently Asked Questions
Common questions about Llama 4 Scout