No — not on this device

DeepSeek V4 Flash 284B-A13B at Q4_K_M needs 160.7 GB against 48.0 GB usable, and the shortfall of 112.7 GB is more than 64 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.

Does not fit 8K context Q4_K_M · 159.7 GB MIT Released 31 Jul 2026 Not in the Ollama library

The V4 for 192–256 GB machines: Q3 squeezes into 192 GB, Q4 wants 256 GB, and 128 GB falls short even at Q2. Cache is modelled as a 576-wide latent; V4 compresses it further at long context, so this is conservative.

Hardware requirements What hardware do I need to run DeepSeek V4 Flash 284B-A13B? →

At Q4_K_M with an 8K context: a 192 GB GPU (Instinct MI300X) or a 256 GB Mac (M3 Ultra). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.

The VRAM budget

weights 159.7 GB
Weights 159.7 GB KV cache @ 8K 0.38 GB Runtime overhead 0.6 GB Over budget 112.7 GB past 48.0 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
Q8_0 281.0 GB 282.0 GB — ~5.2 −0.1% ppl 234.0 GB over
Q6_K 216.9 GB 217.9 GB — ~6.7 −0.4% ppl 169.9 GB over
Q5_K_M 187.5 GB 188.4 GB — ~7.8 −0.8% ppl 140.4 GB over
Q4_K_M 159.7 GB 160.7 GB — ~9.1 −1.9% ppl 112.7 GB over
Q3_K_M 129.3 GB 130.3 GB — ~11 −5.4% ppl 82.3 GB over
Q2_K 110.8 GB 111.7 GB — ~13 −15% ppl 63.7 GB over

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model uses multi-head latent attention, so its cache is a compressed latent rather than full K and V.

How to run it

terminal
$ llama-server \
    -hf deepseek-ai/DeepSeek-V4-Flash-0731:Q4_K_M \
    -c 8192 -ngl 12

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 159.7 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02The GPU shares one pool with the operating system, and these numbers count about 48.0 GB of 64 GB as its share. On Windows the firmware sets the split; on Linux the kernel’s GTT limit does, and can be raised.
03If it falls back to CPU silently, drop the context first, then step down a quantisation.
See all models for this rig Compare with another model

Related models on a Ryzen AI Max+ 395 · 64 GB

ModelTotalTok/s est.
DeepSeek V4.1 Flash 552B-A16B
552B MoE · same family
Q4_K_M · 311.0 GB — Won't fit
DeepSeek V4 Pro 1.6T-A49B
1650B MoE · same family
Q4_K_M · 928.9 GB — Won't fit
GLM-5.3-Flash 320B-A18B
321B MoE · similar size
Q4_K_M · 181.2 GB — Won't fit
Qwen3 235B-A22B
235B MoE · similar size
Q4_K_M · 134.2 GB — Won't fit
Ornith 1.5 397B-A17B
397B MoE · similar size
Q4_K_M · 224.1 GB — Won't fit
Qwen3.5 397B-A17B
403B MoE · similar size
Q4_K_M · 227.4 GB — Won't fit
Qwen3.8-Flash-Next 180B-A6B
180B MoE · similar size
Q4_K_M · 102.0 GB — Won't fit

Every model against the Ryzen AI Max+ 395 · 64 GB →

DeepSeek V4 Flash 284B-A13B on other hardware

DeviceTotalTok/s est.
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
160.7 GB — Won't fit
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
160.7 GB — Won't fit
M4 · 16 GB
10.7 GB usable · 120 GB/s
160.7 GB — Won't fit
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
160.7 GB — Won't fit
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
160.7 GB — Won't fit
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
160.7 GB — Won't fit
Instinct MI300X
190.4 GB usable · 5300 GB/s
160.7 GB 169 Runs great
M3 Ultra · 256 GB
192.0 GB usable · 800 GB/s
160.7 GB 29 Runs great

All 114 devices, by tier →