Ask LLM Advance

Plugin ID: pml_llm

Description

The Ask LLM Advance tool provides access to powerful large language models such as GPT for advanced conversational AI capabilities. This tool supports system messages, chat history management, and temperature control for fine-tuned responses, making it ideal for complex AI interactions and workflows.

Cost Information

• Cost: 4 credits

• PixelML Cost: Based on the selected model pricing

Input Parameters

Parameter	Type	Required	Default	Description
human_message	string	Yes	-	The prompt that is fed to the model. Maximum length: 640,000 characters. Example: "Explain quantum computing"
system_message	string	No	null	System message that tells the model how to behave. Maximum length: 640,000 characters. Example: "You are a helpful assistant"
chat_history_id	string	No	null	The chat history ID used to retrieve previous conversation context. If not provided, a new chat history will be created.
model	string	Yes	-	The AI model to use for generating the response. Available models depend on your PixelML configuration.
temperature	number	No	0.5	Controls response creativity. Higher values (0.7-1.0) = more creative, lower values (0.0-0.3) = more focused. Range: 0.0-1.0

Output

Field	Type	Description
content	string	The generated response content from the selected language model
chat_history_id	string	The chat history ID that can be used to continue the conversation in subsequent requests

How It Works

1. Model Selection: Choose the appropriate AI model for your use case

2. Message Processing: The system message (if provided) sets the model's behavior context

3. History Integration: If a chat history ID is provided, previous conversation context is loaded

4. Response Generation: The model generates a response based on the human message, system instructions, and conversation history

5. History Management: A chat history ID is returned for maintaining conversation continuity

Use Cases

• Conversational AI: Build chatbots and virtual assistants with memory

• Content Generation: Create articles, stories, and marketing copy

• Code Assistance: Get help with programming and technical questions

• Analysis & Research: Analyze data and conduct research with AI assistance

• Creative Writing: Generate creative content with controlled creativity levels

• Customer Support: Provide AI-powered customer service responses

• Educational Tools: Create tutoring and learning assistance applications

Example Usage

Basic Question Answering

Human Message: "What are the benefits of renewable energy?"
Model: "gpt-4"
Temperature: 0.3
System Message: null
Chat History ID: null

Conversational Assistant with Memory

Human Message: "Can you remember what we discussed about solar panels?"
Model: "gpt-3.5-turbo"
Temperature: 0.5
System Message: "You are a knowledgeable environmental consultant."
Chat History ID: "conv_12345"

Creative Writing

Human Message: "Write a short story about time travel"
Model: "gpt-4"
Temperature: 0.8
System Message: "You are a creative science fiction writer."
Chat History ID: null

Temperature Guide

Temperature	Use Case	Behavior
0.0-0.3	Technical Q&A, Analysis	Focused, factual, consistent responses
0.4-0.6	General conversation	Balanced between creativity and accuracy
0.7-1.0	Creative writing, Brainstorming	Highly creative, diverse, unexpected responses

Tips for Best Results

1. Choose the Right Model: Select models based on your task complexity and response quality needs

2. Craft Clear Prompts: Be specific and detailed in your human messages

3. Use System Messages: Define the AI's role and behavior for consistent responses

4. Manage Temperature: Adjust temperature based on whether you need creative or factual responses

5. Maintain Context: Use chat history IDs to maintain conversation continuity across multiple requests

6. Optimize Prompt Length: While 640K characters are supported, shorter, focused prompts often work better

Limitations

• Character Limits: Both human and system messages are limited to 640,000 characters

• Model Availability: Available models depend on your PixelML subscription and configuration

• Context Window: Each model has its own context window limitation

• Rate Limiting: Subject to usage limits based on your plan

• Language Support: Response quality may vary for non-English languages

• Real-time Data: Models don't have access to real-time information unless specifically configured

Privacy & Compliance

• Conversations are processed according to the selected model's privacy policies

• Chat history is stored securely and only accessible via the provided chat history ID

• No conversation data is used for model training without explicit consent

• Complies with AI model provider terms of service

• Users should avoid sharing sensitive personal information in prompts

• Content generation respects copyright and intellectual property guidelines

Error Handling

Common error scenarios:

• Missing Required Fields: Returns error if human_message or model are not provided

• Invalid Model: Returns error for unsupported or unavailable model selections

• Character Limit Exceeded: Returns error if messages exceed 640,000 character limit

• Invalid Temperature: Returns error if temperature is outside 0.0-1.0 range

• Chat History Not Found: Returns error if provided chat_history_id doesn't exist

• Rate Limited: Temporary restriction if usage limits are exceeded

• Model Unavailable: Returns error if selected model is temporarily unavailable