ComfyUI Node
Krea2 SVDQuant Diagnostics
Passthrough node that reports which comfy_kitchen backend the quantized layers dispatch to, plus memory accounting and timings. Paste the output into bug reports.
Krea2 SVDQuant Diagnostics
- model
- model
- report
◄mode▾►
◄tokens4096►
CategoryKrea2/SVDQuant
Inputs (3)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | Output of the Krea2 SVDQuant W4A4 Loader. | |
| mode | COMBO | dispatch: which kernel actually runs (start here). env: versions and memory accounting. bench: quantized vs bf16 per layer. profile: torch.profiler table. compile: graph breaks a TorchCompileModel run would pay. | |
| tokens | INT | 409664–65536 | Sequence length to probe with. 4096 = 1024x1024. Kernel selection is shape-dependent, so match your real run. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| model | MODEL | The model, unchanged - wire it on to your sampler. |
| report | STRING | The report as text, for pasting into a bug report. |