CheckpointLoaderKJ

KJNodes/experimentalExperimental

CheckpointLoaderKJ

Experimental node for patching torch.nn.Linear with CublasLinear.

Experimental: This node is experimental and its behavior may change without notice.

Example

JSON Example

{
  "class_type": "CheckpointLoaderKJ",
  "inputs": {
    "ckpt_name": "https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors",
    "weight_dtype": "default",
    "compute_dtype": "default",
    "patch_cublaslinear": false,
    "sage_attention": "disabled",
    "enable_fp16_accumulation": false
  }
}

This example shows required inputs only. Connection values like ["node_id", 0] should reference actual node IDs from your workflow.

Inputs

Name	Type	Status	Constraints	Default
ckpt_name?	ENUM 0 options URL: Checkpoint	required	-	-
weight_dtype	ENUM 7 options default fp8_e4m3fn fp8_e4m3fn_fast fp8_e5m2 fp16 bf16 fp32	required	-	-
compute_dtype?	ENUM 4 options default fp16 bf16 fp32	required	-	"default"
patch_cublaslinear?	BOOLEAN	required	-	false
sage_attention?	ENUM 8 options disabled auto sageattn_qk_int8_pv_fp16_cuda sageattn_qk_int8_pv_fp16_triton sageattn_qk_int8_pv_fp8_cuda sageattn_qk_int8_pv_fp8_cuda++ sageattn3 sageattn3_per_block_mean	required	-	false
enable_fp16_accumulation?	BOOLEAN	required	-	false

Outputs

Index	Name	Type	Is List	Connection Reference
0	MODEL	MODEL	No	["{node_id}", 0]
1	CLIP	CLIP	No	["{node_id}", 1]
2	VAE	VAE	No	["{node_id}", 2]

How to connect to these outputs

To connect another node's input to an output from this node, use the connection reference format:

["node_id", output_index]

Where node_id is the ID of this CheckpointLoaderKJ node in your workflow, and output_index is the index from the table above.

Example

If this node has ID "5" in your workflow:

• MODEL (MODEL): ["5", 0]
• CLIP (CLIP): ["5", 1]
• VAE (VAE): ["5", 2]

Package: comfyui-kjnodes