Quantization publisher
tencent
tencent publishes 6 quantizations across 2 models in our index, averaging 6.071 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
6
Models covered
2
Avg effective bpw
6.071
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | tencent | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Hy-MT2-7B | Q6_K | 5.74 GiB | unsloth | 5.74 GiB | -0.0% |
| Hy-MT2-7B | Q8_0 | 7.43 GiB | unsloth | 7.43 GiB | -0.0% |
| Hy-MT2-7B | Q4_K_M | 4.31 GiB | unsloth | 4.31 GiB | -0.0% |
| Hy-MT2-7B | Q6_K | 5.74 GiB | mradermacher | 5.74 GiB | -0.0% |
| Hy-MT2-7B | Q8_0 | 7.43 GiB | mradermacher | 7.43 GiB | -0.0% |
| Hy-MT2-7B | Q4_K_M | 4.31 GiB | mradermacher | 4.31 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 |
|---|---|---|
| Hy-MT2-7B | 3 | 4.31 GiB |
| Hy-MT2-1.8B | 3 | 1.06 GiB |