Yes — with 24.9 GB to spare
Qwen3.5 122B-A10B at Q4_K_M fits your Ryzen AI Max+ 395 · 128 GB entirely in unified memory at 8K context, at an estimated 12 tokens per second. There is room for its full 256K window.
Fully in unified memory
8K context
Q4_K_M · 70.3 GB
Apache 2.0
Released 24 Feb 2026
Vision
The 96–128 GB unified-memory model: 122B of knowledge at 10B-active speed.
The VRAM budget
weights 70.3 GB
Weights 70.3 GB
KV cache @ 8K 0.19 GB
Runtime overhead 0.6 GB
Free 24.9 GB of 96.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 123.7 GB | 124.5 GB | — | ~6.7 | −0.1% ppl | 28.5 GB over |
| Q6_K | 95.5 GB | 96.2 GB | — | ~8.7 | −0.4% ppl | 0.2 GB over |
| Q5_K_M | 82.5 GB | 83.3 GB | 256K | 10 | −0.8% ppl | Long context |
| Q4_K_M | 70.3 GB | 71.1 GB | 256K | 12 | −1.9% ppl | Recommended |
| Q3_K_M | 56.9 GB | 57.7 GB | 256K | 15 | −5.4% ppl | Long context |
| Q2_K | 48.7 GB | 49.5 GB | 256K | 17 | −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 12 of its 48 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
$ ollama pull qwen3.5:122b-a10b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run qwen3.5:122b-a10b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
01Download is 70.3 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.