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	🤖 medgemma-27b-it google
Company	google
Release Date	2025-07-09
Parameters (B)	27
Architecture	Decoder-only Transformer
Context Window	128000
Input Modalities	text, image
Open Source	Yes
License	other
Score	0
VRAM (GB)	54
Compute	Single NVIDIA A100 80GB GPU for bf16 inference (~54 GB weights); multimodal variant benefits from H100 80GB or 2x A100 40GB; Q4 quantization (~16 GB) enables single 24GB consumer GPUs
Benchmark: MMLU-Pro	60.2
Pricing (Input)	—
Pricing (Output)	—