What hardware runs Qwen3.5 122B A10B?
Qwen3.5 122B A10B is a 122.11B-parameter mixture-of-experts model with 9.77B active parameters (apache-2.0 license). Arguably the best model a 128 GB Mac can run today. The unified-memory sweet spot the 397B can't reach.
All figures assume an f16 KV cache, a 0.6 GB display reserve on the GPU, and 64 GB of DDR5 system RAM for the offload tiers. Tune these in the calculator.
Minimum VRAM by quant and context
| Quant | File size | 8K | 16K | 32K | 64K | 128K |
|---|---|---|---|---|---|---|
| Q3_K_M * | 52.5 GB | 55.1 GB | 55.9 GB | 57.5 GB | 60.6 GB | 66.8 GB |
| Q4_K_M | 71.3 GB | 73.9 GB | 74.6 GB | 76.2 GB | 79.3 GB | 85.6 GB |
| Q5_K_M | 85.2 GB | 87.8 GB | 88.6 GB | 90.2 GB | 93.3 GB | 99.5 GB |
| Q6_K | 94.1 GB | 96.7 GB | 97.4 GB | 99 GB | 102.1 GB | 108.4 GB |
| Q8_0 | 121.0 GB | 123.5 GB | 124.3 GB | 125.9 GB | 129 GB | 135.3 GB |
Full-GPU figures: weights + f16 KV cache + overhead. * below our recommended floor of Q4_K_M.
GPU compatibility
Fits on GPUExpert offloadPartial offloadCPU only
Quant guidance
Our floor for Qwen3.5 122B A10B is Q4_K_M — below that, quality degrades faster than the VRAM savings are worth (large models tolerate low bpw better than small ones, which is why the floor is lower here). Prefer the highest quant that still lands "Fits on GPU" or “Expert offload” at your context length in the table above.
Recommended run command
Q4_K_M at 32K on a Apple M2 Ultra (128GB)-class GPU (Fits on GPU):
llama-server -m Qwen3.5-122B-A10B-Q4_K_M.gguf -c 32768 --flash-attn -ngl 99
Frequently asked questions
How much VRAM does Qwen3.5 122B A10B need?
At Q4_K_M and 32K context, Qwen3.5 122B A10B needs about 76.2 GB of VRAM to run fully on GPU (weights + KV cache + overhead). As a mixture-of-experts model it can also run with far less VRAM via expert offload, keeping experts in system RAM.
What is the smallest GPU that can run Qwen3.5 122B A10B?
The Apple M2 Ultra (128GB) (96 GB) is the smallest GPU in our set that runs Qwen3.5 122B A10B well at 32K context, using expert offload with 64 GB of system RAM.
What quantization should I use for Qwen3.5 122B A10B?
We recommend Q4_K_M or higher. Q4_K_M weighs 71.3 GB (5.01 bits/weight); going below Q4_K_M costs noticeable quality on a model this size.
How long a context can Qwen3.5 122B A10B handle?
Qwen3.5 122B A10B supports up to 256K tokens. KV cache grows linearly with context.
Can I run Qwen3.5 122B A10B without a GPU?
Not realistically — even at 8K context the weights don't fit in 64 GB of system RAM.