community-verified · measured pool data

Can the RTX 3090 24GB run Qwen3.6-27B-MTP-GGUF?

Yes — community-measured at 49.9 tok/s (16 runs at 4-bit).

Verified single-stream runs on the RTX 3090 24GB from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.

Quantization	Median tok/s	TTFT (ms)	Peak VRAM	Max context tested	Engines
1-bit	49.2	—	2.5 GB	196,608	llama.cpp
4-bit	49.9	465.2	20.2 GB	131,072	llama.cpp

Go deeper

I have a goal — find models that fit my use case → I have hardware — browse everything my rig can run → All machine × model answer pages →