Skip to content

Can RTX 3090 run Qwen3.6 35B-A3B?

Tight: GGUF Q4_K_M needs 23.4 GB at 4K, batch 1; headroom 0.6 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_M23.4TightVerified file
GGUF Q8_039.6Does not fitVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K23.4Tight
16K23.6Tight
32K24.0Tight
128K26.1Does not fit

Generation speed

GGUF Q4_K_M: 523 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs