Quantization publisher

TheDrummer

TheDrummer publishes 24 quantizations across 2 models in our index, averaging 5.436 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
4
Quantizations
24
Models covered
2
Avg effective bpw
5.436
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantTheDrummervsTheirsDifference
Llama-3.3-8B-Instruct-128KQ8_07.95 GiBshb7777.95 GiB0.0%
Llama-3.3-8B-Instruct-128KQ6_K6.14 GiBshb7776.14 GiB0.0%
Llama-3.3-8B-Instruct-128KQ4_K_M4.58 GiBshb7774.58 GiB0.0%

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
Llama-3.3-8B-Instruct-128K62.96 GiB
Mistral-Small-3.2-24B-Instruct-2507188.28 GiB