Yes — with 25.6 GB to spare

Nemotron 3 Super 120B-A12B at Q4_K_M fits your Ryzen AI Max+ 395 · 128 GB entirely in unified memory at 8K context, at an estimated 9.9 tokens per second. There is room for its full 256K window.

Fully in unified memory 8K context Q4_K_M · 69.7 GB NVIDIA Open Model Released Mar 2026 Not in the Ollama library

The open-training-data 120B. Same hybrid layout as Lightning, so 128K context costs under a gigabyte.

What hardware do I need for Nemotron 3 Super 120B-A12B? →

The VRAM budget

weights 69.7 GB
Weights 69.7 GB KV cache @ 8K 0.06 GB Runtime overhead 0.6 GB Free 25.6 GB of 96.0 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
Q8_0 122.7 GB 123.4 GB ~5.6 −0.1% ppl 27.4 GB over
Q6_K 94.7 GB 95.4 GB 89K 7.3 −0.4% ppl Long context
Q5_K_M 81.8 GB 82.5 GB 256K 8.4 −0.8% ppl Long context
Q4_K_M 69.7 GB 70.4 GB 256K 9.9 −1.9% ppl Recommended
Q3_K_M 56.4 GB 57.1 GB 256K 12 −5.4% ppl Long context
Q2_K 48.4 GB 49.0 GB 256K 14 −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. Only 8 of its 88 blocks keep a per-token KV cache; the rest are linear-attention, Mamba or convolution blocks with a fixed-size state.

How to run it

terminal
$ llama-server \
    -hf nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16:Q4_K_M \
    -c 8192 -ngl 99

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

01Download is 69.7 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 256K context on this card.
See all models for this rig Compare with another model