Quantization publisher
ekampra
ekampra publishes 11 quantizations across 1 models in our index, averaging 9.153 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
1
Quantizations
11
Models covered
1
Avg effective bpw
9.153
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | ekampra | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q5_K_S | 15.53 GiB | DevQuasar-9 | 15.48 GiB | +0.3% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q5_K_M | 16.00 GiB | DevQuasar-9 | 15.96 GiB | +0.3% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q3_K_L | 11.73 GiB | DevQuasar-9 | 11.68 GiB | +0.4% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q3_K_M | 10.83 GiB | DevQuasar-9 | 10.79 GiB | +0.4% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q3_K_S | 9.76 GiB | DevQuasar-9 | 9.72 GiB | +0.5% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q4_K_M | 13.65 GiB | DevQuasar-9 | 13.61 GiB | +0.3% |
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | Q4_K_S | 12.84 GiB | DevQuasar-9 | 12.80 GiB | +0.4% |
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 |
|---|---|---|
| Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B | 11 | 9.76 GiB |