ComfyUI Node
LM Studio Query
Query LM Studio for text generation with optional image input, model management, and multi-seed prompt generation.
LM Studio Query
- images
- generated_texts
- model_ref
◄system_promptYou are a helpful assistant.►
◄user_promptDescribe this image in detail.►
◄model_name►
◄server_urlhttp://localhost:1234►
◄temperature0.70►
◄max_tokens256►
◄top_p0.90►
◄seed0►
◄prompts_number1►
◄unload_after_usefalse►
Categorypirog/ai
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| system_prompt | STRING | You are a helpful assistant. | System prompt to set the AI's behavior and role |
| user_prompt | STRING | Describe this image in detail. | User prompt for the query. Use with images for descriptions or standalone for text generation. |
| model_name | STRING | Model name or partial name to search for. Leave empty to use currently loaded model. | |
| server_url | STRING | http://localhost:1234 | LM Studio server URL |
| temperature | FLOAT | 0.700–2 | Sampling temperature (0.0 = deterministic, 2.0 = very random) |
| max_tokens | INT | 2561–4096 | Maximum number of tokens to generate |
| top_p | FLOAT | 0.900–1 | Nucleus sampling parameter |
| seed | INT | 00–18446744073709550000 | Random seed for reproducible results. 0 = random each time. |
| prompts_number | INT | 11–10 | Number of prompts to generate with different seeds |
| unload_after_use | BOOLEAN | false | Unload the model after processing to free memory |
| imagesopt | IMAGE | Optional images for visual queries (batch processing supported) |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| generated_texts | STRING | Array of generated text responses |
| model_ref | STRING | Reference to the loaded model name |