DeepSeek System Requirements for Web, App and Local Use

“Will this laptop run DeepSeek?” is the wrong question, because there are five different ways to use it and each has its own answer. The DeepSeek system requirements you need depend entirely on whether you are opening chat.deepseek.com in a browser, installing the mobile app, calling the API from a server, running an R1 distill in Ollama, or self-hosting DeepSeek’s full open weights on a GPU cluster. The web and mobile clients ask very little of your hardware. Self-hosting the full models asks for data-centre GPUs. This guide gives you the requirements for each path, based on DeepSeek’s official documentation, model cards and app-store listings as of October 2026.

The five ways to use DeepSeek — and what each needs

Before listing specs, decide which surface you actually need. The hardware bill goes up by several orders of magnitude from the lightest to the heaviest option.

The first three paths offload compute to DeepSeek’s infrastructure. Only the last two are real “system requirements” questions in the hardware sense. We cover all five below.

Web chat system requirements

The web client at chat.deepseek.com is a thin browser app. The model runs on DeepSeek’s servers, not yours — according to DeepSeek’s 17 October 2026 Change Log, the official web and app models were updated to DeepSeek-V4.1-Flash-1017 that day.

Browser

• Chrome, Edge, Firefox, Safari or Brave — current major version
• JavaScript enabled, cookies enabled (login uses session cookies)
• Stable internet (a few hundred kbps is enough for text)

Operating system

• Any desktop operating system that runs a current version of one of those browsers — Windows, macOS, Linux or ChromeOS
• No GPU, no minimum RAM beyond what your browser already consumes

Model and mode options in the web interface change over time — DeepSeek’s 13 August 2026 announcement, for example, said V4 Pro was available on the app and web via “Expert Mode”. None of these options changes what your own device needs. For a deeper walkthrough of the interface, see our DeepSeek chat guide.

Mobile app system requirements

The official iOS and Android apps are also thin clients. Any reasonably recent phone that meets the store’s minimum OS version will run them.

Store requirements change with app updates, so check the listing before you install on an older device. If you are checking that you have the genuine app rather than a clone, our guide to verifying the official DeepSeek app walks through the publisher fields.

API system requirements

The DeepSeek API is OpenAI-compatible at https://api.deepseek.com and Anthropic-compatible at https://api.deepseek.com/anthropic. Chat requests hit POST /chat/completions, the OpenAI-compatible endpoint. The API is stateless — your client must resend the conversation history on every request. This is the opposite of the web app, which keeps your chat history for you.

System requirements for calling the API are negligible: any machine with a current Python or Node.js runtime (or just curl) that can reach https://api.deepseek.com over HTTPS. All inference runs on DeepSeek’s side, whichever model you pick — as of October 2026 the API serves deepseek-flash (DeepSeek-V4.1-Flash-1017) and deepseek-v4-pro (DeepSeek-V4-Pro-0813), both with a 1M-token context. For model IDs, parameters and migration notes, see our DeepSeek API documentation.

Local deployment: distilled R1 models on Ollama

If you want a model on your own hardware, the realistic starting point is one of the DeepSeek-R1 distilled checkpoints — smaller dense models trained to mimic R1’s reasoning. The distilled variants range from 1.5B to 70B parameters and are installable in one command via Ollama. Licensing depends on the base model each distill was built from, so check the model’s Hugging Face card if licensing matters to you.

VRAM requirements by distill size (Q4 quantization)

The VRAM figures are approximate and depend on the quantisation variant and context length. The 32B distill at Q4_K_M uses around 20 GB VRAM — a fit for an RTX 3090 or RTX 4090 with about 4 GB of headroom for the KV cache. The 70B distill at Q4 needs about 40 GB, beyond any single consumer GPU, so the usual options are two 24 GB cards or a high-memory Mac. Throughput varies widely with GPU, quantisation and prompt length, so benchmark on your own hardware before you buy.

For Apple Silicon, unified memory is the trick: the GPU shares the system RAM pool, so an M3 Max or M4 Max with 64 GB or 128 GB can load models that would crash a typical gaming PC, at lower throughput. Inference is meaningfully slower than on a 4090 but the capacity advantage is real. See our DeepSeek on Mac guide for the specifics.

Minimum CPU and RAM

• Modern x86-64 CPU (Intel 10th gen / AMD Ryzen 3000 or newer) or Apple Silicon
• 16 GB system RAM is the practical floor; 32 GB recommended once you go past 7B
• NVMe SSD with 50–200 GB free for model weights and KV cache scratch
• Linux, Windows 10/11 with WSL2, or macOS 13+

The fastest path is Ollama. Once installed, a single line pulls and runs the model:

ollama run deepseek-r1:32b

Step-by-step setup, including GPU offloading and Open-WebUI, is in our running DeepSeek on Ollama tutorial. For a fuller installation walkthrough across operating systems, see install DeepSeek locally.

Local deployment: full DeepSeek-V4 weights

This is the heavy end. The two models behind DeepSeek’s current API publish open weights under the MIT license on Hugging Face: DeepSeek-V4.1-Flash-1017 (served as deepseek-flash) and DeepSeek-V4-Pro-0813 (served as deepseek-v4-pro). Neither model card, as read on 23 October 2026, publishes a minimum-hardware table, so size your cluster from the checkpoint files listed in each repository and the serving recipes they link to. Hardware estimates written for the April 2026 V4 preview checkpoints do not carry over to these newer releases.

DeepSeek-V4.1-Flash-1017

The V4.1-Flash-1017 card points to its inference folder for weight conversion and local inference, and recommends temperature 1.0, top_p 1.0, a 1M-token context window and max_tokens of at least 256K. Long contexts are what drive memory beyond the weights themselves: DeepSeek’s card for V4.1-Flash (the September 2026 release that 1017 updates) reports a KV-cache footprint of 890 bytes per token, roughly a quarter of the earlier V4-Flash. By that figure, one full 1M-token sequence needs about 0.89 GB of KV cache, multiplied by the number of long sequences you serve at once.

Realistic requirements for serving the full Flash weights:

• A multi-GPU data-centre server — size it from the repository’s checkpoint files plus KV-cache headroom
• Fast NVMe storage with room for the checkpoint and runtime files
• A current release of vLLM or SGLang for inference; vLLM exposes an OpenAI-compatible API by default
• A recent CUDA toolkit and NVIDIA driver, with NCCL configured for multi-GPU
• Linux (Ubuntu LTS is the most common target)

DeepSeek-V4-Pro-0813

V4-Pro-0813 is built on the V4-Pro (Preview) model structure with an added DSpark speculative-decoding module, according to its model card. The card’s vLLM example serves the model on a single node with four NVIDIA GB300 GPUs (data parallel across four GPUs, with expert parallelism), its SGLang example uses tensor parallelism across four GPUs, and both link to the vLLM recipe and SGLang cookbook for other hardware configurations. That puts Pro out of reach for almost everyone except teams with access to current data-centre GPU nodes, either owned or rented from a cloud provider. If data sovereignty is not a hard requirement, the API is usually the cheaper route at moderate volumes.

BF16 / full-precision is a different problem

DeepSeek ships these checkpoints in mixed low precision. If you intend to run the model at full BF16 precision instead, plan for several times more GPU memory, plus headroom for KV cache, activation buffers and real batch sizes. For most teams this is the moment to stop and price out the API instead: as of October 2026, deepseek-flash lists $0.30 per 1M cache-miss input tokens and $1.20 per 1M output tokens at peak, and $0.15 / $0.60 off-peak (official pricing page). Our DeepSeek API pricing page lays out every rate, and the DeepSeek pricing calculator lets you run your own numbers.

Recommendation by user profile

• Casual user / researcher. Web or app. Zero hardware investment.
• Indie developer. API on deepseek-flash. Pay per token from a prepaid balance. No GPUs to babysit.
• Privacy-sensitive solo developer. R1-Distill-Qwen-32B on a single RTX 4090 or used 3090 via Ollama. Good reasoning, fully offline, no per-token bill.
• Small team, regulated industry. Self-hosted DeepSeek-V4.1-Flash-1017 on a multi-GPU data-centre server under vLLM or SGLang. Real upfront cost, but you own the stack.
• Frontier-tier workloads. Stay on the API for deepseek-v4-pro until your daily volume justifies a multi-GPU node of your own.

For deeper guidance on choosing between hosted and local, see our DeepSeek free vs paid comparison and the broader DeepSeek beginner guides hub.