Intern-S2 Models — Hardware Requirements

3 Intern-S2 models from InternLM and the community, from the smallest that runs in 72.6 GB of VRAM up to 404.1B parameters. Every row links to full quantization tables and GPU compatibility.

All Intern-S2 Models by Size

ModelParamsContext
Intern S2 Preview36.1B262K
Intern S2 397B403.4B262K
Intern S2 Preview 397B404.1B262K

Frequently Asked Questions

How much VRAM do I need to run a Intern-S2 model?
The smallest Intern-S2 model, Intern S2 Preview, runs from 72.6 GB of VRAM at an aggressive quantization. Larger family members need proportionally more — see the table above for every model.
Which Intern-S2 models can I run on a 16 GB GPU?
No Intern-S2 model currently fits in 16 GB of VRAM — the family starts at 72.6 GB.
What is the most popular Intern-S2 model to run locally?
Intern S2 397B is the most downloaded Intern-S2 model in local-friendly quantized formats. It runs from 111.4 GB of VRAM.
Intern-S2 Models — VRAM & Hardware Requirements | llmrun