Quantization publisher
RachidAR
RachidAR publishes 12 quantizations across 4 models in our index, averaging 9.446 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 7 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
4
Quantizations
12
Models covered
4
Avg effective bpw
9.446
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | RachidAR | vs | Theirs | Difference |
|---|---|---|---|---|---|
| gemma-4-26B-A4B-it-assistant | Q4_K_M | 0.31 GiB | AtomicChat | 0.30 GiB | +1.3% |
| gemma-4-26B-A4B-it-assistant | Q5_K_M | 0.32 GiB | AtomicChat | 0.32 GiB | +0.7% |
| gemma-4-26B-A4B-it-assistant | Q8_0 | 0.43 GiB | Radamanthys11 | 0.43 GiB | 0.0% |
| gemma-4-31B-it-qat-q4_0-unquantized-assistant | BF16 | 0.89 GiB | Gaboo | 0.89 GiB | 0.0% |
| gemma-4-12B-it-qat-q4_0-unquantized-assistant | Q8_0 | 0.43 GiB | Janvitos | 0.43 GiB | -0.0% |
| gemma-4-26B-A4B-it-assistant | Q8_0 | 0.43 GiB | AtomicChat | 0.43 GiB | -0.0% |
| gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant | Q8_0 | 0.43 GiB | Janvitos | 0.43 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
| Model | Quantizations | Smallest |
|---|---|---|
| gemma-4-12B-it-qat-q4_0-unquantized-assistant | 1 | 0.43 GiB |
| gemma-4-26B-A4B-it-assistant | 7 | 0.30 GiB |
| gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant | 3 | 0.30 GiB |
| gemma-4-31B-it-qat-q4_0-unquantized-assistant | 1 | 0.89 GiB |