Can you run Qwen 3.5 9B on Radeon RX 6600?
Short answer
Yes, but only just — Qwen 3.5 9B leaves 1.6 GB spare on Radeon RX 6600.
- Model file
- 5.3 GB
- Needs in VRAM, with context
- about 6.4 GB
- Card memory
- 8 GB
- Measured on this hardware
- not measured
How this was worked out
Qwen 3.5 9B is a 5.3 GB file. Running it needs the weights in memory plus room for the context window and the runtime, which comes to roughly 6.4 GB in practice. Radeon RX 6600 has 8 GB, which leaves 1.6 GB. That is arithmetic rather than a benchmark: it tells you whether the model loads, not how it feels.
What we have actually measured
We have not timed Qwen 3.5 9B on Radeon RX 6600. The closest figure we have is 27 tokens per second on RTX 3060 12 GB, Vulkan backend. Rather than scale that number to your card and present the result as a benchmark, this page leaves it out. The method is published at /benchmarks if you would like to run it yourself.
If it is tight
1.6 GB is not much of a margin. Your desktop compositor and anything with a video playing are competing for the same memory, so this one loads but can fall back to the CPU under pressure. If it stutters, drop to the next model down. Nightjar sizes the context window to the memory it finds, so a smaller card gets a shorter window rather than a failed load.
Smaller models that fit the same card
- Qwen 3.5 4B on Radeon RX 6600 — 2.6 GB
- Gemma 4 E2B-it on Radeon RX 6600 — 3.2 GB
- Gemma 4 E4B-it on Radeon RX 6600 — 5 GB
Related
- Up a levelCan your PC run a local AI model?
- Closely relatedCan you run Gemma 4 12B-it on Radeon RX 6600?
- Closely relatedCan you run Gemma 4 26B-A4B-it on Radeon RX 6600?
- Closely relatedCan you run Gemma 4 31B-it on Radeon RX 6600?
- Closely relatedCan you run Gemma 4 E2B-it on Radeon RX 6600?
- Elsewhere on the siteLocal AI on Radeon RX 6600: which models run and how fast
- Elsewhere on the siteQwen 3.5 9B in Nightjar: speed, size and what it is good at