What are the hardware requirements for DeepSeek V4.1 Flash 1017 NVFP4?

DeepSeek V4.1 Flash 1017 NVFP4 requires approximately 315.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed DeepSeek V4.1 Flash 1017 NVFP4 and how large is it?

DeepSeek V4.1 Flash 1017 NVFP4 is developed by nvidia, a model with 664.9b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate DeepSeek V4.1 Flash 1017 NVFP4?

DeepSeek V4.1 Flash 1017 NVFP4 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

DeepSeek V4.1 Flash 1017 NVFP4 by nvidia — VRAM 315.6GB

Name: DeepSeek V4.1 Flash 1017 NVFP4
Author: nvidia

DeepSeek V4.1 Flash 1017 NVFP4 is an open-source language model by nvidia. Features: 664.9b LLM, VRAM: 315.6GB, License: mit.

8-bit Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-... Deepseek v41 Deepseekv4 Fp4 Fp8 Image-text-to-text Model optimizer Modelopt Nvidia Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/nvidia/DeepSeek-V4.1-Flash-1017-NVFP4

DeepSeek V4.1 Flash 1017 NVFP4 Parameters and Internals

LLM Name	DeepSeek V4.1 Flash 1017 NVFP4
Repository 🤗	https://huggingface.co/nvidia/DeepSeek-V4.1-Flash-1017-NVFP4
Base Model(s)	deepseek-ai/DeepSeek-V4.1-Flash-1017 deepseek-ai/DeepSeek-V4.1-Flash-1017
Model Size	664.9b
Required VRAM	315.6 GB
Updated	2026-11-02
Maintainer	nvidia
Model Type	deepseek_v41
Model Files	1.0 GB: 1-of-48 1.3 GB: 2-of-48 7.8 GB: 3-of-48 7.8 GB: 4-of-48 7.8 GB: 5-of-48 7.8 GB: 6-of-48 7.8 GB: 7-of-48 7.8 GB: 8-of-48 7.8 GB: 9-of-48 7.8 GB: 10-of-48 7.8 GB: 11-of-48 7.8 GB: 12-of-48 7.8 GB: 13-of-48 7.8 GB: 14-of-48 7.8 GB: 15-of-48 7.8 GB: 16-of-48 7.8 GB: 17-of-48 7.8 GB: 18-of-48 7.8 GB: 19-of-48 7.8 GB: 20-of-48 7.8 GB: 21-of-48 7.8 GB: 22-of-48 7.8 GB: 23-of-48 7.8 GB: 24-of-48 7.8 GB: 25-of-48 7.8 GB: 26-of-48 7.8 GB: 27-of-48 7.8 GB: 28-of-48 7.8 GB: 29-of-48 7.8 GB: 30-of-48 7.8 GB: 31-of-48 7.8 GB: 32-of-48 7.8 GB: 33-of-48 7.8 GB: 34-of-48 7.8 GB: 35-of-48 7.8 GB: 36-of-48 7.8 GB: 37-of-48 7.8 GB: 38-of-48 7.8 GB: 39-of-48 7.8 GB: 40-of-48 7.8 GB: 41-of-48 7.8 GB: 42-of-48 1.3 GB: 43-of-48
Model Architecture	DeepseekV41ForCausalLM
License	mit
Transformers Version	5.6.0

Rank the DeepSeek V4.1 Flash 1017 NVFP4 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to DeepSeek V4.1 Flash 1017 NVFP4

Best Alternatives	Context / RAM	Downloads	Likes
DeepSeek V4.1 Flash 1017 NVFP4	0K / 290.9 GB	668	20

Note: green Score (e.g. "73.2") means that the model is better than nvidia/DeepSeek-V4.1-Flash-1017-NVFP4.