ComfyUI Node
LM Studio Chat
A ComfyUI node in LM Studio with 9 inputs and 3 outputs.
LM Studio Chat
- model
- images
- options
- responses
- queries
- conversation_name
◄conversation_choiceNew Conversation►
◄conversation_name►
◄modeone-by-one►
◄system_prompt►
◄user_prompt►
◄temp_convofalse►
CategoryLM Studio
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| model | LMSTUDIO_MODEL | Model settings including model_id, temperature, max_tokens, seed, and image_max_size. | |
| conversation_choice | COMBO | New Conversation | Pick an existing conversation or select 'New Conversation' to start a new one. Provide 'conversation_name' to name it, or leave blank to auto-generate. |
| conversation_name | STRING | If provided, creates/uses this conversation. If left blank, uses the dropdown selection. | |
| mode | COMBO | one-by-one | Batch sends all images in a single request; one-by-one sends one request per image. |
| system_prompt | STRING | System instructions for the assistant's behavior. If the conversation has no prior messages, this will be set as the initial system message. | |
| user_prompt | STRING | User message appended to the conversation and sent to the model on this run. | |
| imagesopt | IMAGE | Optional images to include with the user message. Resized to image_max_size. | |
| optionsopt | LMSTUDIO_OPTIONS | Per-request overrides (temperature, max_tokens, seed, top_p, top_k, penalties, stop). These take precedence over values from the Model node. | |
| temp_convoopt | BOOLEAN | false | Store/load conversation in a temporary workspace (cleared on startup). New conversations use temp when enabled. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| responses | STRING | — |
| queries | STRING | — |
| conversation_name | STRING | — |