RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen3.5 9B Base run on the RX 7900 XT?

Yes. At 8K context the best fit is Q8_0, using about 10.0 GB of the 19.0 GB usable on the RX 7900 XT. Estimated speed ceiling: ~84 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •4.6 GB Fits4.7 GB Fits5.4 GB Fits8.4 GB Fits209
3-bit (Q3_K_M) •5.3 GB Fits5.4 GB Fits6.2 GB Fits9.2 GB Fits173
4-bit (Q4_K_M) •6.2 GB Fits6.4 GB Fits7.1 GB Fits10.1 GB Fits142
5-bit (Q5_K_M) •7.0 GB Fits7.1 GB Fits7.9 GB Fits10.9 GB Fits124
6-bit (Q6_K) •7.9 GB Fits8.0 GB Fits8.7 GB Fits11.7 GB Fits109
8-bit (Q8_0) •9.9 GB Fits10.0 GB Fits10.7 GB Fits13.7 GB Fits84
16-bit (F16/BF16) •17.7 GB Tight17.8 GB Tight18.6 GB Tight21.6 GB No45

• real GGUF file size from mradermacher/Qwen3.5-9B-Base-GGUF. Speed is a bandwidth ceiling (800 GB/s), not a benchmark. Methodology.

More

Qwen3.5 9B Base on every GPU and Mac · Every model that runs on the RX 7900 XT · Other models that fit 24 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen3.5-9B-Base), AMD specs. Data as of 2026-11-17.