GPU VRAM calculator

Run DeepSeek-V4-Flash-0731 locally: hardware requirements

DeepSeek-V4-Flash-0731 needs about 165 GB of GPU memory for one person at 32K context, and about 5.4 GB more for each extra person at 64K. For 80 users working as AI agents, plan on 248 GB, which takes 2 × H200 NVL.

DeepSeek, China. From its files on Hugging Face, checked 2026-11-05.

Context per person: more context lets the model work with longer conversations and more files, and needs more GPU memory for every person. As a guide: 8K for short chat, 32K for chat over documents, 64K for AI agents, whose instructions, tools and files all count.

80 users means about 16 people working at the same moment. That share is a planning assumption; if you know yours, choose the total that matches it.

Work it out for your team: deepseek-ai/DeepSeek-V4-Flash-0731

On Hugging Face

DeepSeek, China · 304B parameters · License: mit

What you need for 16 active users

2 × H200 NVL

It holds the model and up to 19 active users at 64K context.

The server

512 GB RAM
48 CPU cores
1 TB NVMe

Rules of thumb

GPU memory in total
248 GB
Weights
167 GB (published files)
Conversation cache, all active
86.0 GB
Engine (vLLM)
3.0 GB

Ways to need less

Your options by card

80 users means about 16 using the model at any moment. The others are not waiting; they are simply doing something else. If more become active than a build handles, they queue. 9 cards too small for this model are left out.

CardMemoryActive users, one serverFor your 16 active
A10080 GB264 × A100
B200180 GB332 × B200
H10080 GB264 × H100
H200 NVL141 GB192 × H200 NVL
L40S48 GB206 × L40S
RTX 6000 Ada48 GB36 servers, each with 4 × RTX 6000 Ada
RTX PRO 5000 Blackwell48 GB36 servers, each with 4 × RTX PRO 5000 Blackwell
RTX PRO 6000 Blackwell96 GB203 × RTX PRO 6000 Blackwell

These are memory limits. A card can hold more people than it can answer quickly, so test your own workload before you buy. More than one card in a server needs an engine that splits the model.

DeepSeek-V4-Flash-0731 by team size

AI agents, 1 user in 5 active at a time, 64K context, as published.

Total usersAt onceGPU memoryWhat you need
204183 GB2 × H200 NVL
8016248 GB2 × H200 NVL
12024291 GB2 × B200
500100699 GB5 × B200

All models

How this is worked out

GPU memory = weights + a conversation cache for each person + the serving engine, from this model's files on Hugging Face. These are memory limits, not speed. The full method and its limits.

Questions

How much GPU memory does DeepSeek-V4-Flash-0731 need?

DeepSeek-V4-Flash-0731 needs about 165 GB of GPU memory for one person at 32K context, and about 5.4 GB more for each extra person at 64K. For 80 users working as AI agents, plan on 248 GB, which takes 2 × H200 NVL.

How many active users can DeepSeek-V4-Flash-0731 handle on one 96 GB GPU?

DeepSeek-V4-Flash-0731 does not fit on one 96 GB card such as the RTX PRO 6000 Blackwell at 64K context. Compare larger cards or a smaller precision above.

Can DeepSeek-V4-Flash-0731 run on a 24 GB GPU such as the RTX 4090?

Not as published for one person at 32K context: it needs about 165 GB. A 4-bit or 8-bit build may fit, if the model has one.

What license does DeepSeek-V4-Flash-0731 use?

Hugging Face lists its license as mit. Check the license on the model's page before you use it.

Running a model on your own servers?

NeueCode is private autonomous intelligence for enterprise: it takes on your organization's tasks on your own GPUs, with nothing leaving your network, a person approving every action, and all of it on record. It runs each model on one card, so read the single-card figures above.

What you need to run NeueCode