ComfyUI Node
LM Loader
Picks the chat model a wrapper node runs (Captioner, Prompter): the family, its model and precision, one LoRA and its strength, and whether the model stays loaded. It loads nothing: lm describes the model, and the node it feeds finds the file (or downloads it on first use), reads the LoRA and loads them when it runs. The lists follow the selected family and model.
LM Loader
- lm
◄familyQwen LM►
◄modelQwen3.5-9B►
◄precisionINT8 ConvRot►
◄loraNone►
◄lora_strength1.00►
◄keep_model_loadedtrue►
CategoryBCNodes/lm
Inputs (6)
| Name | Type | Default | Description |
|---|---|---|---|
| family | COMBO | Qwen LM | The LM family: the catalog, folder and chat templates the model comes from. The model, precision and lora lists show the selected family's; a value of another family is an error that names this family's. |
| model | COMBO | Qwen3.5-9B | The chat model (the built-in ones all see images): the built-in list plus the models.yaml in models/Qwen-LM. The file is looked for by name in models/Qwen-LM (subfolders too), then anywhere under models/text_encoders, and downloaded into models/Qwen-LM when missing. |
| precision | COMBO | INT8 ConvRot | The model file's weights: BF16 (full), INT8 ConvRot (8-bit, about half the size), W4A8 (4-bit, Qwen3.8-27B only). Only the selected model's precisions are shown; another one is an error. |
| lora | COMBO | None | A LoRA from models/Qwen-LM/loras: a .safetensors file (PEFT or ComfyUI key names) or a PEFT folder (adapter_config.json + adapter_model.safetensors). Its alpha is read from the file, its metadata or its config file (a file in ComfyUI's names without one: alpha = rank, core's convention); a LoRA trained on another base model is refused. Only the selected family's LoRAs are shown. |
| lora_strength | FLOAT | 1.00-100–100 | How strongly the LoRA changes the model: strength x alpha / rank x B @ A, core's LoRA math. 0 runs the model without it. |
| keep_model_loaded | BOOLEAN | true | On: the model (with its LoRA) stays loaded after the node that runs it, for the next run; one LM model is held at a time, shared with the other LM nodes. Off: that node unloads it once when its call ends, after an error too. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| lm | BC_LM | The model for a wrapper node's lm input (Captioner, Prompter): a description, nothing is loaded here. |