Compare LLM Models

Select 2-5 models to compare side by side.

Select models to compare 0 / 5 selected

Top downloads

🤖 DeepSeek-V4-Pro-0813 deepseek-ai 🤖 GLM-5.3 zai-org 🤖 Kimi-K3 moonshotai

Select 2-5 models

Attribute	🤖 Llama-4-Scout-17B-16E-Instruct meta-llama
Company	meta-llama
Release Date	2025-04-05
Parameters (B)	109
Architecture	MoE Transformer
Context Window	10000000
Input Modalities	text, image
Open Source	Yes
License	other
Score	10
VRAM (GB)	80
Compute	Single NVIDIA H100 80GB with on-the-fly INT4 quantization; ~218GB VRAM for full BF16 weights
Benchmark: GPQA	46.5
Benchmark: MBPP	67.8
Benchmark: MMLU-Pro	74.3
Pricing (Input)	—
Pricing (Output)	—