deepseek-ai
DeepSeek V4.1 Flash (1017)
1M-context mixture-of-experts model with 8-16B active parameters for reasoning, coding and agents; updated weights from 2026-10-17.image text to textmit312.3K downloadsHugging Face
No server in stock right now has the 614 GB this model needs for inference.
GPU memory it needs
From model cards and vendor documentation. Lower precision needs less memory, with slightly lower quality. Estimates are marked.
| What you want to do | Precision | Minimum | Recommended | Setup | Basis |
|---|---|---|---|---|---|
| Run the model | FP8, 8-bit | 614 GB | 768 GB | 4x GB200 (vLLM recipe default, one NVL4 tray) | Published |