Hugging Face
Send Hugging Face output to Pushcut
Generate text from a Hugging Face language model in response to your prompt.
Developer info
- API endpoint slug
- hugging_face.run_text_generation
- Runtime method
- performQuery("hugging_face.run_text_generation"}
Query fields
Model Text input > Other
- Label
- Model
- Helper text
- The Hugging Face model ID of a conversational model, e.g. mistralai/Mistral-7B-Instruct-v0.2. Requires Inference Providers access on your account.
- Slug
- model_id
- Required
- true
- Can have default value
- true
Prompt Text input
- Label
- Prompt
- Helper text
- The prompt to send to the model.
- Slug
- prompt
- Required
- true
- Can have default value
- true
Max new tokens Text input > Other
- Label
- Max new tokens
- Helper text
- Maximum number of tokens to generate. Defaults to 200.
- Slug
- max_new_tokens
- Required
- false
- Can have default value
- true
Ingredients
Generated text The text generated by the model.
- Slug
- generated_text
- Filter code
- HuggingFace.runTextGeneration.GeneratedText
- Type
- String
- Example
- The capital of France is Paris.
Model The model that produced the response.
- Slug
- model_id
- Filter code
- HuggingFace.runTextGeneration.ModelId
- Type
- String
- Example
- mistralai/Mistral-7B-Instruct-v0.2
Prompt used The prompt that was sent to the model.
- Slug
- prompt_used
- Filter code
- HuggingFace.runTextGeneration.PromptUsed
- Type
- String
- Example
- What is the capital of France?
