Quantization publisher
RobinsonLabs
RobinsonLabs publishes 38 quantizations across 4 models in our index, averaging 3.927 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 2 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
4
Quantizations
38
Models covered
4
Avg effective bpw
3.927
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | RobinsonLabs | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Qwen3.5-99B | Q4_K_M | 56.06 GiB | 0xSero | 56.06 GiB | 0.0% |
| Qwen3.5-99B | Q6_K | 75.74 GiB | 0xSero | 75.74 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 |
|---|---|---|
| Qwen3.5-99B | 10 | 27.51 GiB |
| Qwen3.5-88B | 10 | 24.44 GiB |
| Qwen3.5-REAP-212B-A17B | 8 | 52.41 GiB |
| Qwen3.5-REAP-262B-A17B | 10 | 64.50 GiB |