Quantization publisher

batiai

batiai publishes 4 quantizations across 2 models in our index, averaging 7.275 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 6 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
2
Quantizations
4
Models covered
2
Avg effective bpw
7.275
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantbatiaivsTheirsDifference
Qwen3-Embedding-4BQ8_03.99 GiBcstr3.99 GiB0.0%
Qwen3-VL-Embedding-8BQ6_K5.79 GiBVesNFF5.79 GiB0.0%
Qwen3-VL-Embedding-8BQ6_K5.79 GiBmradermacher5.79 GiB-0.0%
Qwen3-VL-Embedding-8BQ8_07.50 GiBmradermacher7.50 GiB-0.0%
Qwen3-Embedding-4BQ6_K3.08 GiBmradermacher3.08 GiB-0.0%
Qwen3-Embedding-4BQ8_03.99 GiBmradermacher3.99 GiB-0.0%

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
Qwen3-VL-Embedding-8B25.79 GiB
Qwen3-Embedding-4B23.08 GiB