Yes — with 25.5 GB to spare

Ornith 1.5 35B-A3B at Q4_K_M fits your Radeon Pro W7900 entirely on the GPU at 8K context, at an estimated 108 tokens per second. There is room for its full 256K window.

Fully on GPU 8K context Q4_K_M · 20.2 GB MIT Released 19 Aug 2026 New this week

A reasoning-first MIT build on the Qwen3.6 35B-A3B architecture (thinks before every answer). Same VRAM as its base.

What hardware do I need for Ornith 1.5 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 25.5 GB of 46.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	35.5 GB	36.3 GB	256K	62	−0.1% ppl	Long context
Q6_K	27.4 GB	28.2 GB	256K	80	−0.4% ppl	Long context
Q5_K_M	23.7 GB	24.5 GB	256K	92	−0.8% ppl	Long context
Q4_K_M	20.2 GB	20.9 GB	256K	108	−1.9% ppl	Recommended
Q3_K_M	16.3 GB	17.1 GB	256K	134	−5.4% ppl	Long context
Q2_K	14.0 GB	14.8 GB	256K	156	−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 ornith-1.5:35b
$ OLLAMA_CONTEXT_LENGTH=8192 \
    ollama run ornith-1.5:35b

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