ComfyUI Node
M8 · Local LLM Inference
Runs inference on a local GGUF model: no network, no API key.Models live in ComfyUI's models/LLM folder.Pair it with an mmproj file in the same folder to enable vision; a skill node can supply a knowledge pack.
M8 · Local LLM Inference
- image
- skill
- text
- status
◄model▾►
◄system_prompt►
◄user_prompt►
◄extra_text►
◄mmproj▾►
◄max_tokens512►
◄temperature0.70►
◄top_p0.95►
◄ctx4096►
◄gpu_layers-1►
◄thinkingfalse►
CategoryM8/LLM
Inputs (13)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | The main .gguf in models/LLM. If the list is empty, check that folder. | |
| system_prompt | STRING | System prompt. Combined with any skill into the system block, skill first. | |
| user_prompt | STRING | What you want to ask this round. | |
| extra_text | STRING | Extra text appended after the prompt. Good for additional requirements, style constraints, fields to reference. | |
| imageopt | IMAGE | Images for the model to look at. Without it the run is text only. Needs an mmproj file next to the model, otherwise vision does nothing. | |
| skillopt | M8_SKILL | Wire the output of a skill loader node here to use it as a knowledge pack. | |
| mmprojopt | COMBO | Multimodal projection file. Auto-paired by name; pick one manually if that guessed wrong. | |
| max_tokensopt | INT | 51216–8192 | — |
| temperatureopt | FLOAT | 0.700–2 | — |
| top_popt | FLOAT | 0.950.05–1 | — |
| ctxopt | INT | 4096512–32768 | Context length. Larger uses more memory; only needed for long conversations or many images. |
| gpu_layersopt | INT | -1-1–200 | -1 = offload as many layers as the GPU allows (recommended; depends on whether llama-cpp-python was built with CUDA). 0 = CPU only. |
| thinkingopt | BOOLEAN | false | Whether the model should show its thinking. Off gives cleaner output; some models ignore it, in which case ask in the prompt too. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |
| status | STRING | — |