ComfyUI Node
Mistral AI completion
Makes a single call to the Mistral AI API. Set the API key in API-key.txt
Mistral AI completion
- image
- response
◄model▾►
◄temperature1.0►
◄top_p0.9►
◄max_tokens512►
◄presence_penalty0.0►
◄frequency_penalty0.0►
◄promptYour response is directly used as a prompt: avoid quotation marks, brackets, etc! Focus on visually percievable descriptors, not touchy feely fluff. Give me a single SD prompt for the following:
►
◄context—►
◄random_seed—►
Categoryutils/text
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | 9 options: pixtral-large-latest, pixtral-12b-latest, ministral-3b-latest, ministral-8b-latest, ministral-14b-latest, open-mistral-nemo, +3 | |
| temperature | FLOAT | 1.00–1.5 | — |
| top_p | FLOAT | 0.90–1 | — |
| max_tokens | INT | 512 | — |
| presence_penalty | FLOAT | 0.0-2–2 | — |
| frequency_penalty | FLOAT | 0.0-2–2 | — |
| prompt | STRING | Your response is directly used as a prompt: avoid quotation marks, brackets, etc! Focus on visually percievable descriptors, not touchy feely fluff. Give me a single SD prompt for the following: | — |
| contextopt | STRING | Use Mistral AI prompt loader node to supply the context for few-shot prompting. | |
| imageopt | IMAGE | Only Pixtral models support image input. | |
| random_seedopt | INT | If you don't supply a fixed seed, it's randomized on the server. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| response | STRING | — |