Quantization publisher
stepfun-ai
stepfun-ai publishes 7 quantizations across 1 models in our index, averaging 6.198 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
7
Models covered
1
Avg effective bpw
6.198
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | stepfun-ai | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Step-3.7-Flash | IQ3_XXS | 70.56 GiB | bartowski | 78.38 GiB | -10.0% |
| Step-3.7-Flash | Q3_K_L | 95.46 GiB | bartowski | 88.94 GiB | +7.3% |
| Step-3.7-Flash | Q4_K_S | 103.84 GiB | bartowski | 109.05 GiB | -4.8% |
| Step-3.7-Flash | Q8_0 | 195.04 GiB | bartowski | 197.44 GiB | -1.2% |
| Step-3.7-Flash | IQ4_XS | 97.78 GiB | bartowski | 99.94 GiB | -2.2% |
| Step-3.7-Flash | Q3_K_M | 87.36 GiB | bartowski | 85.51 GiB | +2.2% |
| Step-3.7-Flash | Q8_0 | 195.04 GiB | unsloth | 195.04 GiB | -0.0% |
| Step-3.7-Flash | BF16 | 366.96 GiB | unsloth | 366.96 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 |
|---|---|---|
| Step-3.7-Flash | 7 | 70.56 GiB |