|
<< Click to Display Table of Contents >> Navigation: 3. Script Language > AI - Artificial Intelligence Commands > AID. - DeepSeek > !Chat Operation > AID. - Artificial Intelligence DeepSeek Utility |
MiniRobotLanguage (MRL)
AID.AskT
Query the DeepSeek API with Plain Text Response
Intention
AskT Command: Simple AI Interaction
The AID.AskT command queries the DeepSeek API and returns a plain text response, perfect for direct display or lightweight applications.
It leverages DeepSeek’s cost-effective API (e.g., $0.27/1M input tokens and $1.10/1M output tokens for DeepSeek Chat) for efficient, human-readable AI output.
It’s part of the AID - DeepSeek API suite, optimized for MiniRobotLanguage workflows.
The AID.AskT command sends a prompt to the DeepSeek API, sets the response format to text via AID_SetRespFormat "text", and returns a plain text response processed by HTP_AnalyzeDeepSeekResponse.
It supports storing the response in a variable and optionally copying it to the clipboard, ideal for quick, readable outputs.
The AID.AskT command is essential for:
•Simplicity: Provides plain text for immediate use or display without parsing.
•Cost Efficiency: Matches AID.Ask’s low cost ($1.10/1M output tokens vs. GPT-4o’s $15.00/1M).
•User Interaction: Perfect for chatbots or interfaces needing human-readable responses.
Supply a prompt (P1), an optional variable for the text response (P2), and an optional clipboard flag (P3).
Supports 'deepseek-chat' ($0.27/$1.10 per 1M tokens) and 'deepseek-reasoner' ($0.55/$2.19 per 1M tokens).
Use this command for direct, readable output rather than structured data processing.
Cost Model and Usage
Billing is token-based: 'deepseek-chat' at $0.27/1M input and $1.10/1M output; 'deepseek-reasoner' at $0.55/1M input and $2.19/1M output (February 2025).
A 100-input, 200-output token query costs ~$0.00025 for Chat, suitable for real-time user interfaces or low-overhead tasks.
Example Usage
AID.Set Model|deepseek-chat
AID.Set Temperature|1.0
AID.AskT|Write a haiku|$$HAI|CLIP
DBP.Haiku: $$HAI
Sets a creative temperature, requests a haiku, stores the text (e.g., "Cherry blossoms fall...") in $$HAI, and copies it.
AID.AskT|What is 2 + 2?|$$ANS
DBP.Answer: $$ANS
Stores the text response (e.g., "4") in $$ANS for direct display.
Illustration
┌────────────────────┐
│ Prompt: "Write a │
│ short story" │
├──────┬─────────────┤
│ Text │ "Once upon a│
│ Resp │ time..." │
├──────┼─────────────┤
│ Cost │ ~$0.0005 │
└──────┴─────────────┘
Flow of a prompt yielding a text response with estimated cost.
Syntax
AID.AskT|P1[|P2][|P3]
Parameter Explanation
P1 - Mandatory prompt sent to the DeepSeek API.
P2 - Optional variable to store the text response.
P3 - Optional flag (e.g., "CLIP") to copy the response to the clipboard.
Example
AID.Set Model|deepseek-reasoner
AID.AskT|Explain gravity|$$EXP|CLIP
DBP.Explanation: $$EXP
ENR.
Uses 'deepseek-reasoner' to explain gravity, storing the text (e.g., "Gravity is the force...") in $$EXP.
Comparison: AID.Ask vs. AID.AskT
Use AID.AskT for plain text output suitable for direct display or minimal processing (e.g., user interfaces, logging).
Choose AID.Ask for JSON output when parsing or structured data is required. AID.AskT is lighter and faster for simple text needs.
Remarks
- Returns "DeepSeek did not answer!" if the API fails.
- Response depends on prior settings (e.g., temperature, max tokens).
Limitations
- Requires a valid API key and endpoint.
- Limited by max tokens (default 1024).
See also:
• AID.Ask