Quantization publisher
ddh0
ddh0 publishes 8 quantizations across 1 models in our index, averaging 4.319 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 8 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
1
Quantizations
8
Models covered
1
Avg effective bpw
4.319
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | ddh0 | vs | Theirs | Difference |
|---|---|---|---|---|---|
| L3.3-Electra-R1-70b | Q5_K_M | 46.52 GiB | bartowski | 46.52 GiB | -0.0% |
| L3.3-Electra-R1-70b | Q4_K_M | 39.60 GiB | bartowski | 39.60 GiB | 0.0% |
| L3.3-Electra-R1-70b | Q4_K_S | 37.58 GiB | bartowski | 37.58 GiB | 0.0% |
| L3.3-Electra-R1-70b | Q5_K_S | 45.32 GiB | bartowski | 45.32 GiB | 0.0% |
| L3.3-Electra-R1-70b | IQ2_M | 22.46 GiB | bartowski | 22.46 GiB | 0.0% |
| L3.3-Electra-R1-70b | IQ3_M | 29.74 GiB | bartowski | 29.74 GiB | 0.0% |
| L3.3-Electra-R1-70b | IQ4_XS | 35.30 GiB | bartowski | 35.30 GiB | 0.0% |
| L3.3-Electra-R1-70b | IQ3_XS | 27.29 GiB | bartowski | 27.29 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 |
|---|---|---|
| L3.3-Electra-R1-70b | 8 | 22.46 GiB |