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

Written by Jakub Rusinowski · Last updated August 15, 2026

Yes

Yes — Devstral-2 123B at Q6_K needs about 104.5 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB) (~23.5 GB spare), at ~5.5 tok/s (estimated), with room for about 65,536 tokens of context.

Confidence: medium · Recommended quantization: Q6_K · Estimated speed: ~5.5 tok/s

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

Usable memory for models128 GB
Memory bandwidth256 GB/s
Form factorMini PC
Operating systemWindows or Linux
Memory upgradeableNo — soldered

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

QuantMemory neededFits 128 GB?Max contextEst. speedDownload
F16249.7 GB✗ No246 GB
Q8_0134.4 GB✗ No130.7 GB
Q6_K104.5 GB✓ Yes64K~5.5 tok/s100.9 GB
Q5_K_M90.9 GB✓ Yes64K~6.4 tok/s87.2 GB
Q4_K_M77.9 GB✓ Yes64K~7.4 tok/s74.3 GB
Q3_K_M56.1 GB✓ Yes64K~10.1 tok/s52.4 GB
Q2_K44.1 GB✓ Yes64K~12.8 tok/s40.4 GB

Which Devstral sizes fit

VariantNeeds at Q4_K_MFits?Est. speed
Devstral-2 123B77.9 GB✓ Fits~7.4 tok/s
Devstral Small 24B16.6 GB✓ Fits~12.3 tok/s
Devstral-2 22B15.7 GB✓ Fits~13.2 tok/s

What to watch out for

Framework Desktop 128 GB limitations

Recommended setup

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

How these numbers are calculated

FAQ

Can I run Devstral on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?

Yes — Devstral-2 123B at Q6_K needs about 104.5 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB) (~23.5 GB spare), at ~5.5 tok/s (estimated), with room for about 65,536 tokens of context.

Which quantization of Devstral should I use on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?

Q6_K — it needs about 104.5 GB of the 128 GB available, downloads as roughly 100.9 GB, and runs at an estimated 5.5 tokens/sec with up to 64K of context.

What limits Devstral on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?

Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.

Which runtime should I use?

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

Other Computers

Other Models on Framework Desktop (Ryzen AI Max+ 395, 128 GB)

Devstral on GPUs

What This Model Is Good At

Model & Tools

← Can I Run It? | Devstral model page | Check your hardware