community pool / model answer

Spark-X2.5-4B

chat · XHToken/Spark-X2.5-4B

84.7 tok/s · model median · 5 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
RTX 3060 12GB	3-bit	76.1 tok/s	reported	171	3.8 GB	4,096	llama.cpp
RTX 3060 12GB	4-bit	90.7 tok/s	reported	161	3.5 GB	4,096	llama.cpp
RTX 3060 12GB	5-bit	89.2 tok/s	reported	163	4 GB	4,096	llama.cpp
RTX 3060 12GB	6-bit	77.9 tok/s	reported	175	4.6 GB	4,096	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.