Unsloth
Qwen3.5 122B A10B (122B parameters) requires approximately 75.9 GB of VRAM with Q3_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 88 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run Qwen3.5 122B A10B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "unsloth/Qwen3.5-122B-A10B-GGUF" \
--hf-file "Qwen3.5-122B-A10B-GGUF-Q3_K_M.gguf" \
-c 4096 -ngl 99Quick specs
Related models
Inference speed
Estimated decode speed (tokens/sec) for Qwen3.5 122B A10B at Q3_K_M across popular GPUs and Apple Silicon, including multi-GPU rigs, using the fastest local runtime per device. Fastest is MacBook Pro M4 Max 128GB at ~9 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
MacBook Pro M4 Max 128GB | 128 GB | Q3_K_M | 9.3 | Offloads |
Mac Studio M3 Ultra 256GB | 256 GB | Q3_K_M | 8.7 | Fits |
Mac Studio M2 Ultra 128GB | 128 GB | Q3_K_M | 7.2 | Offloads |
Mac Studio M1 Ultra 128GB | 128 GB | Q3_K_M | 6.8 | Offloads |
2× RX 7900 XTX 24GB | 48 GB | Q3_K_M | 4.7 | Too big |
MacBook Pro M4 Max 64GB | 64 GB | Q3_K_M | 4.5 | Too big |
 RTX 5090 32GB | 32 GB | Q3_K_M | 2.8 | Too big |
MacBook Pro M4 Pro 48GB | 48 GB | Q3_K_M | 2.6 | Too big |
| 2× RTX 4090 24GB | 48 GB | Q3_K_M | 2.5 | Too big |
| 2× RTX 3090 24GB | 48 GB | Q3_K_M | 2.3 | Too big |
| RTX 4090 24GB | 24 GB | Q3_K_M | 2.0 | Too big |
| RTX 4080 Super 16GB | 16 GB | Q3_K_M | 2.0 | Too big |
| RTX 3090 24GB | 24 GB | Q3_K_M | 2.0 | Too big |
| RTX 4070 12GB | 12 GB | Q3_K_M | 2.0 | Too big |
| RTX 3060 12GB | 12 GB | Q3_K_M | 2.0 | Too big |
| RTX 4060 8GB | 8 GB | Q3_K_M | 2.0 | Too big |
RX 7900 XTX 24GB | 24 GB | Q3_K_M | 2.0 | Too big |
MacBook Pro M3 Max 64GB | 64 GB | Q3_K_M | 2.0 | Too big |
MacBook Pro M1 Max 64GB | 64 GB | Q3_K_M | 2.0 | Too big |
| 4× RTX 3060 12GB | 48 GB | Q3_K_M | 2.0 | Too big |
Estimates for single-stream decoding at Q3_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
Quick picks
Best hardware
Run this model
Quantization
How much VRAM Qwen3.5 122B A10B (122B) needs at each GGUF quant, and whether it fits a 24 GB card (RTX 4090 / 3090).
| Quant | Bits | VRAM (weights) | Quality | Fits 24 GB? |
|---|---|---|---|---|
| Q2_K | 2 | 47.6 GB | Low | Too big |
| Q3_K_S | 3 | 59.8 GB | Low | Too big |
| NVFP4 | 4 | 68.3 GB | Medium | Too big |
| Q4_K_M | 4 | 74.4 GB | Medium | Too big |
| Q5_K_M | 5 | 87.8 GB | High | Too big |
| Q6_K | 6 | 100 GB | High | Too big |
| Q8_0 | 8 | 130.5 GB | Very High | Too big |
| F16 | 16 | 250.1 GB | Maximum | Too big |
VRAM shown is quantized weights only; add ~1–3 GB runtime overhead plus KV cache for your context length. Lower quants trade quality for memory — Q4_K_M is the usual sweet spot; Q2/Q3 only when you must fit a bigger model.
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
Qwen3.5 122B A10B (122B parameters) requires approximately 75.9 GB of VRAM with Q3_K_M quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, Mac Studio M3 Ultra 256GB can run Qwen3.5 122B A10B with a compatibility score of 47/100. It provides 256 GB of memory and achieves approximately 8.7 tokens per second.
The recommended quantization for Qwen3.5 122B A10B is Q3_K_M, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for Qwen3.5 122B A10B: AMD Instinct MI300A 128GB (score: 55/100), NVIDIA H200 141GB (score: 55/100), NVIDIA H200 PCIe 141GB (score: 55/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, Qwen3.5 122B A10B is well-suited for chat. It was designed with these use cases in mind.
See also
