Yes — with 74.7 GB to spare
LLM-jp 4 33B Thinking at Q4_K_M fits your Ryzen AI Max+ 395 · 128 GB entirely in unified memory at 8K context, at an estimated 7.7 tokens per second. There is room for its full 64K window.
Fully in unified memory
8K context
Q4_K_M · 18.7 GB
Apache 2.0
Released 14 Aug 2026
New this week
Not in the Ollama library
Japan’s national-institute reasoning model, Japanese and English. A plain dense Llama-style 33B: Q4 is a tight 24 GB fit.
The VRAM budget
weights 18.7 GB
Weights 18.7 GB
KV cache @ 8K 2.00 GB
Runtime overhead 0.6 GB
Free 74.7 GB of 96.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 32.9 GB | 35.5 GB | 64K | 4.4 | −0.1% ppl | Long context |
| Q6_K | 25.4 GB | 28.0 GB | 64K | 5.6 | −0.4% ppl | Long context |
| Q5_K_M | 21.9 GB | 24.5 GB | 64K | 6.5 | −0.8% ppl | Long context |
| Q4_K_M | 18.7 GB | 21.3 GB | 64K | 7.7 | −1.9% ppl | Recommended |
| Q3_K_M | 15.1 GB | 17.7 GB | 64K | 9.5 | −5.4% ppl | Long context |
| Q2_K | 12.9 GB | 15.5 GB | 64K | 11 | −15% ppl | Long context |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.
How to run it
$ llama-server \
-hf llm-jp/llm-jp-4-33b-thinking:Q4_K_M \
-c 8192 -ngl 99
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
01Download is 18.7 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02macOS caps what the GPU may wire down at about 96.0 GB of 128 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to the model's full 64K context on this card.