ComfyUI Node
HSWQ Model Loader (Validation & VRAM Estimation)
Validates HSWQ model loading configuration and estimates VRAM usage. This node does NOT actually load models. It performs: - Validation of model_path, quantization_bit, and device - Estimation of VRAM usage based on model size and bit width - Reports pass/fail with reasons For actual model loading, use the corresponding GPU-enabled node.
HSWQ Model Loader (Validation & VRAM Estimation)
- valid
- vram_usage
- reason
◄model_path►
◄quantization_bit4►
◄devicecuda►
◄model_size_gb6.5►
CategoryParametric/Models
Inputs (4)
| Name | Type | Default | Description |
|---|---|---|---|
| model_path | STRING | Model path. | |
| quantization_bit | COMBO | 4 | Quantization bit. One of 4 choices. Default 4. |
| device | COMBO | cuda | Device. One of 2 choices. Default cuda. |
| model_size_gbopt | FLOAT | 6.50–100 | Model size gb. From 0 to 100. Default 6.5. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| valid | BOOLEAN | — |
| vram_usage | FLOAT | — |
| reason | STRING | — |