community pool / model answer

granite-4.2-30b-GGUF

chat · ibm-granite/granite-4.2-30b-GGUF

29.7 tok/s · model median · 109 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
BC-250 8GB	2-bit	16.2 tok/s	measured	18,681	- GB	4,357	llama.cpp
CMP 170HX 64GB	4-bit	31.4 tok/s	measured	27,292	- GB	67,589	llama.cpp
CMP 170HX 64GB	8-bit	27.5 tok/s	measured	27,293	- GB	67,588	llama.cpp
CMP 170HX 64GB	16-bit	56.2 tok/s	measured	2,625	- GB	65,800	vllm

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.