Quantization publisher

prometheusAIR

prometheusAIR publishes 4 quantizations across 1 models in our index, averaging 3.977 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 3 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
1
Quantizations
4
Models covered
1
Avg effective bpw
3.977
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantprometheusAIRvsTheirsDifference
Solar-Open2-250BQ2_K88.87 GiBvcruz30585.18 GiB+4.3%
Solar-Open2-250BIQ1_M56.30 GiBvcruz30552.67 GiB+6.9%
Solar-Open2-250BIQ4_XS126.88 GiBvcruz305124.46 GiB+1.9%

A quantization label describes a target, not a recipe. Publishers make different choices about which tensors to keep at higher precision, and some apply an importance matrix while others don't — so two files both honestly labelled the same thing can differ measurably in size and in quality.

Models they publish

ModelQuantizationsSmallest
Solar-Open2-250B456.30 GiB