Yes — with 90.3 GB to spare

Fara 7B at Q4_K_M fits your Ryzen AI Max+ 395 · 128 GB entirely in unified memory at 8K context, at an estimated 31 tokens per second. There is room for its full 125K window.

Fully in unified memory 8K context Q4_K_M · 4.7 GB MIT Released 24 Nov 2025 Vision Not in the Ollama library

A web computer-use agent on a Qwen2.5-VL base — it clicks, fills forms and stops for permission. Not a chat model; size it like an 8B with vision.

What hardware do I need for Fara 7B? →

The VRAM budget

weights 4.7 GB
Weights 4.7 GB KV cache @ 8K 0.44 GB Runtime overhead 0.6 GB Free 90.3 GB of 96.0 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
F16 15.4 GB 16.5 GB 125K 9.3 Reference Long context
Q8_0 8.2 GB 9.2 GB 125K 17 −0.1% ppl Long context
Q6_K 6.3 GB 7.4 GB 125K 23 −0.4% ppl Long context
Q5_K_M 5.5 GB 6.5 GB 125K 26 −0.8% ppl Long context
Q4_K_M 4.7 GB 5.7 GB 125K 31 −1.9% ppl Recommended
Q3_K_M 3.8 GB 4.8 GB 125K 38 −5.4% ppl Long context
Q2_K 3.2 GB 4.3 GB 125K 44 −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

terminal
$ llama-server \
    -hf microsoft/Fara-7B:Q4_K_M \
    -c 8192 -ngl 99

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

01Download is 4.7 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 125K context on this card.
See all models for this rig Compare with another model