Nodes/ComfyUI-Venice-API/Generate Text Advanced BETA (Venice)
ComfyUI Node

Generate Text Advanced BETA (Venice)

Text Generation node that makes use of Venice.AI's text generation API. Use Textgen Parameters (Venice) node to pass on extra Venice.AI specific parameters.Does not have chat history context.

By DraconicDragon·Created 2 years ago·Updated 7 months ago· 7
Generate Text Advanced BETA (Venice)
  • image_for_vision
  • response
modelllama-3.1-405b
prompt
system_prompt
enable_system_prompttrue
frequency_penalty0.00
presence_penalty0.00
repetition_penalty1.20
max_temp1.50
min_temp0.10
max_completion_tokens420
temperature0.50
top_k40
top_p0.80
min_p0.05
enable_visionfalse
venice_parameters
Categoryvenice.ai

Inputs (17)

NameTypeDefaultDescription
modelCOMBOllama-3.1-405bThe model to use for text generation.
promptSTRINGThe prompt to generate text from. Ask, command or chat with the model.
system_promptSTRINGOptional system prompt to guide the model's behavior.
enable_system_promptBOOLEANtrueEnable or disable system prompt being passed on.
frequency_penaltyFLOAT0.00-2–2Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.
presence_penaltyFLOAT0.00-2–2Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.
repetition_penaltyFLOAT1.200–21.0 means no penalty. Values > 1.0 discourage repetition.
max_tempFLOAT1.500–2Maximum temperature value for dynamic temperature scaling.
min_tempFLOAT0.100–2Minimum temperature value for dynamic temperature scaling.
max_completion_tokensINT4201–131072An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.
temperatureFLOAT0.500–2Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both.
top_kINT40The number of highest probability vocabulary tokens to keep for top-k-filtering.
top_pFLOAT0.800–2An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
min_pFLOAT0.050–1Sets a minimum probability threshold for token selection. Tokens with probabilities below this value are filtered out.
enable_visionBOOLEANfalseEnable or disable vision tasks. Requires image_for_vision input to be populated and for the LLM to actually support vision tasks to process.
venice_parametersoptSTRINGOptional input. Use the Textgen Parameters (Venice) node to use extra, venice specific parameters for text generation.
image_for_visionoptIMAGEOptional input. Add an image for vision-supported LLMs to process. Will only be processed if 'enable_vision' is 'True' and if the LLM actually supports vision tasks.

Outputs (1)

NameTypeDescription
responseSTRING