Autor: Jakub Rusinowski · Ostatnia aktualizacja: 24 kwietnia 2026
Model library → DeepSeek V4 → DeepSeek V4-Flash
284B-total MoE with 13B active parameters per token, 1M context, MIT license. Even at Q4 this needs roughly 140-160GB depending on quant — realistically a 2-4x RTX 4090/5090 workstation with heavy CPU offload, or a single 80GB+ datacenter GPU. Local inference is experimental: only community llama.cpp forks (e.g. antirez's branch, ~18 tok/s on a 96GB RTX 6000) run it today; no mainline llama.cpp or stable local Ollama build yet. Treat as 'enthusiast workstation, WIP software' rather than a turnkey local model.
DeepSeek V4-Flash needs about 172 GB of VRAM at Q4 (experimental) — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
| Parameters | 284 Billion (13B active) |
| Context window | 1,000,000 |
| Architecture | Mixture-of-Experts |
| Provider | DeepSeek |
| Licence | MIT |
| Specified at | Q4 (experimental) |
| System RAM | 256 GB |
| Record updated | 2026-04-24 |
MIT — commercial use permitted. Commercial use permitted. No usage restrictions beyond attribution.
Modelled on a reference NVIDIA RTX 4090 (24 GB), with no KV cache (this record has no published architecture). Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.
| Quant | Weights | VRAM needed | Est. speed | Fit on 24 GB |
|---|---|---|---|---|
| Q2_K | 93.4 GB | 94.2 GB | — | Won't fit |
| Q3_K_M | 121.1 GB | 121.9 GB | — | Won't fit |
| Q4_K_M | 171.5 GB | 172.3 GB | — | Won't fit |
| Q5_K_M | 201.3 GB | 202.1 GB | — | Won't fit |
| Q6_K | 232.9 GB | 233.7 GB | — | Won't fit |
| Q8_0 | 301.8 GB | 302.6 GB | — | Won't fit |
| F16 | 568.0 GB | 568.8 GB | — | Won't fit |
Want the memory numbers alone, at every quantization level and your own context length? Use the DeepSeek V4-Flash VRAM calculator.
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
The cheapest catalogued GPU that runs DeepSeek V4-Flash is the Apple M2 Ultra (192 GB).
Install Ollama, then run:
ollama run deepseek-v4-flash (cloud-hosted on Ollama; local = WIP llama.cpp forks only)Weights on Hugging Face: deepseek-ai/DeepSeek-V4-Flash.
Best for: reasoning, coding, long context, enthusiast workstation.
← All DeepSeek V4 models | VRAM calculator | Check your own hardware