Yes — with 169.5 GB to spare

Qwen3.6 35B-A3B at Q4_K_M fits your Instinct MI300X entirely on the GPU at 8K context, at an estimated 665 tokens per second. There is room for its full 256K window.

Fully on GPU 8K context Q4_K_M · 20.2 GB Apache 2.0 Released 16 Apr 2026 Vision

Mixture of experts with ~3B active: the fastest serious model a 24 GB card runs, and the best MoE under 40B on agentic coding.

What hardware do I need for Qwen3.6 35B-A3B? →

The VRAM budget

weights 20.2 GB

Weights 20.2 GB KV cache @ 8K 0.16 GB Runtime overhead 0.6 GB Free 169.5 GB of 190.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	35.5 GB	36.3 GB	256K	378	−0.1% ppl	Long context
Q6_K	27.4 GB	28.2 GB	256K	490	−0.4% ppl	Long context
Q5_K_M	23.7 GB	24.5 GB	256K	567	−0.8% ppl	Long context
Q4_K_M	20.2 GB	20.9 GB	256K	665	−1.9% ppl	Recommended
Q3_K_M	16.3 GB	17.1 GB	256K	822	−5.4% ppl	Long context
Q2_K	14.0 GB	14.8 GB	256K	959	−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 10 of its 40 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 llama.cpp LM Studio vLLM

terminal

$ ollama pull qwen3.6:35b-a3b
$ OLLAMA_CONTEXT_LENGTH=8192 \
    ollama run qwen3.6:35b-a3b

The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.

01Download is 20.2 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.

02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.

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