Skip to content

Can Radeon RX 9060 XT 8G run Qwen3.5 9B?

Comfortable: GGUF Q4_K_M needs 6.6 GB at 4K, batch 1; headroom 1.4 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q4_K_M6.6ComfortableVerified file
GGUF Q8_010.4Does not fitVerified file
GGUF Q5_K_M7.6TightVerified file
GGUF Q6_K8.5Does not fitVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K6.6Comfortable
16K7.0Comfortable
32K7.5Tight
128K10.8Does not fit

Generation speed

GGUF Q4_K_M: 41–55 tok/s · Derived range

Other model / GPU pairs