Can I Run DeepSeek V4 on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?

Autor: Jakub Rusinowski · Ostatnia aktualizacja: 24 kwietnia 2026

Yes, but it is tight

Yes, but it is tight — DeepSeek V4-Flash at Q3_K_M needs about 122.4 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB), leaving only ~5.6 GB before the runtime starts swapping. Expect ~30.7 tok/s (estimated), with room for about 65,536 tokens of context.

Confidence: low · Recommended quantization: Q3_K_M · Estimated speed: ~30.7 tok/s

Buy This HardwareApple Mac Studio M4 Max — 128 GB VRAM · 35 W board power Deploy in the Cloud NowRunPod

or compare on Vast.ai

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.

Framework Desktop (Ryzen AI Max+ 395, 128 GB) — what it gives a model

Usable memory for models	128 GB
Memory bandwidth	256 GB/s
Form factor	Mini PC
Operating system	Windows or Linux
Memory upgradeable	No — soldered

DeepSeek V4 on Framework Desktop (Ryzen AI Max+ 395, 128 GB): memory by quantization

Quant	Memory needed	Fits 128 GB?	Max context	Est. speed	Download
F16	569.4 GB	✗ No	—	—	568 GB
Q8_0	303.1 GB	✗ No	—	—	301.8 GB
Q6_K	234.3 GB	✗ No	—	—	232.9 GB
Q5_K_M	202.7 GB	✗ No	—	—	201.3 GB
Q4_K_M	172.8 GB	✗ No	—	—	171.5 GB
Q3_K_M	122.4 GB	✓ Yes	64K	~30.7 tok/s	121.1 GB
Q2_K	94.7 GB	✓ Yes	256K	~38.5 tok/s	93.4 GB

Which DeepSeek V4 sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
DeepSeek V4-Pro	967.4 GB	✗ Too large	—
DeepSeek V4-Flash	172.8 GB	✗ Too large	—

What to watch out for

Q3_K_M is the only quantization that fits, and it is a heavily degraded one — expect noticeably worse output than the same model at Q4_K_M. A smaller model at Q4 is usually the better trade.
2 larger variants of DeepSeek V4 do not fit and would need CPU offload or different hardware.
This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
Memory on this machine is not upgradeable, so the configuration you buy is the ceiling for every model you will ever run on it.

Framework Desktop 128 GB limitations

Memory is soldered LPDDR5X — unusually for Framework, this is the one component that cannot be upgraded.
ROCm/Vulkan support for Strix Halo is younger than CUDA; check your runtime supports it before buying.

Recommended setup

Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving

How these numbers are calculated

Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
128 GB unified memory at 256 GB/s, shared between CPU and GPU.
Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
KV cache is inferred from the parameter count because this model does not publish its full attention configuration.

FAQ