Meta
Can the Ryzen AI Max+ 395 (64 GB) run Llama 3.3 70B?
We earn from qualifying purchases via affiliate links. Disclosure.
Yes. Llama 3.3 70B is a 43 GB download (Q4_K_M). The Ryzen AI Max+ 395 (64 GB) gives the GPU 48 GB, leaving about 5 GB for context and runtime (our arithmetic).
Does it fit?
| Llama 3.3 70B memory | 43 GB at Q4_K_M (Ollama: llama3.3 tags) |
|---|---|
| Ryzen AI Max+ 395 (64 GB) GPU memory | 64 GB unified memory, of which AMD says up to 48 GB can be dedicated graphics memory through Variable Graphics Memory. We use 48 GB for fit. (AMD) |
| Left over | About 5 GB (our arithmetic) |
| Verdict | Fits 4 GB or more to spare (our estimate bands) |
Recommended quantization and context
Use the Q4_K_M build that Ollama's tag downloads; the memory figure above is for that build. The model supports up to 128K of context, but longer context uses more memory. Ollama starts at a 4096-token context by default (Ollama FAQ).
With about 5 GB spare you have room to raise the context. Raise it in steps and re-check with ollama ps.
On this mini PC the GPU's share of memory depends on settings. On Windows, set Variable Graphics Memory in AMD Software (Performance, then Tuning); on Linux, see AMD's ROCm notes on BIOS VRAM and GTT. Details are in the install guide.
Run it
ollama run llama3.3:70bThen check it's on the GPU:
ollama psThe PROCESSOR column should read 100% GPU. Driver setup for AMD cards on Windows, Linux and macOS, plus common fixes, is in the install and run section.
Where to buy the Ryzen AI Max+ 395 (64 GB)
Minisforum MS-S1 MAX 64GB / 2TB (Ryzen AI Max+ 395)
$2,599 Minisforum · checked Oct 27, 2026, 12:40 PM CT
Listed as available in the store data
Check price at Minisforum (opens in a new tab)FAQ
Can the Ryzen AI Max+ 395 (64 GB) run Llama 3.3 70B?
Yes: Llama 3.3 70B needs 43 GB (Q4_K_M, Ollama library) and the Ryzen AI Max+ 395 (64 GB) gives the GPU 48 GB, leaving about 5 GB. That leaves room to raise the context above the 4096-token default. The fit bands are our estimate.
How much VRAM does Llama 3.3 70B need?
43 GB for the Q4_K_M build in the Ollama library, plus roughly 1–4 GB for context and runtime (our estimate). See the 48 GB page for other models that fit.
What command runs Llama 3.3 70B?
ollama run llama3.3:70b. Then run "ollama ps" and check the PROCESSOR column reads 100% GPU.
How fast is Llama 3.3 70B on the Ryzen AI Max+ 395 (64 GB)?
We don't list a speed because we have no sourced measurement for this pairing. Speed depends on memory bandwidth, quantization and context length.