community pool / model answer

Qwen3.5-9B-GGUF

chat · unsloth/Qwen3.5-9B-GGUF

51.9 tok/s · model median · 15 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
RTX 3060 12GB	4-bit	52.2 tok/s	reported	2,492	6.5 GB	32,768	llama.cpp
RTX 3060 12GB ×2	4-bit	51 tok/s	measured	5,640	7.3 GB	131,072	llama.cpp
RTX 3090 24GB ×2	4-bit	110.1 tok/s	reported	-	- GB	8,192	llama.cpp
RTX 5070 Ti 16GB	4-bit	124.1 tok/s	reported	-	- GB	8,192	llama.cpp
Tesla P100-PCIE-16GB 16GB	5-bit	32.4 tok/s	reported	-	- GB	2,048	llama.cpp
Tesla P100-PCIE-16GB 16GB ×2	5-bit	49.9 tok/s	reported	-	- GB	2,048	llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.