ComfyUI Node
GROQ LLM (Advanced)
A ComfyUI node in ComfyGroq/Text with 15 inputs and 1 output.
GROQ LLM (Advanced)
- context_documents
- response
◄api_key►
◄modelllama3-70b-8192►
◄prompt►
◄system_messageYou are a helpful AI assistant with deep knowledge and analytical skills.►
◄conversation_history►
◄temperature0.70►
◄max_tokens1024►
◄top_p0.90►
◄frequency_penalty0.0►
◄presence_penalty0.0►
◄seed-1►
◄format_outputmarkdown►
◄enable_function_callingfalse►
◄stop_sequences►
CategoryComfyGroq/Text
Inputs (15)
| Name | Type | Default | Description |
|---|---|---|---|
| api_key | STRING | Your GROQ API key. Leave empty to use GROQ_API_KEY environment variable. | |
| model | COMBO | llama3-70b-8192 | LLaMA 3 70B: Meta's 70B parameter model, great for complex tasks |
| prompt | STRING | Your input prompt or question | |
| system_message | STRING | You are a helpful AI assistant with deep knowledge and analytical skills. | System message to guide the model's behavior |
| conversation_history | STRING | Previous conversation history (format: 'user: message assistant: response') | |
| temperature | FLOAT | 0.700–2 | Lower values make output more deterministic, higher more creative |
| max_tokens | INT | 10241–8192 | Maximum number of tokens to generate |
| top_p | FLOAT | 0.900.1–1 | Nucleus sampling: consider only the top tokens with cumulative probability >= top_p |
| frequency_penalty | FLOAT | 0.0-2–2 | Positive values penalize new tokens based on existing frequency |
| presence_penalty | FLOAT | 0.0-2–2 | Positive values penalize new tokens based on whether they appear in the text so far |
| seed | INT | -1-1–4294967295 | Random seed (-1 for random) |
| format_output | COMBO | markdown | Format for the output |
| enable_function_calling | BOOLEAN | false | — |
| context_documentsopt | LIST | — | |
| stop_sequencesopt | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| response | STRING | — |