The AI Engine Behind Onelist
OpenAI powers multiple Onelist agents with GPT models for content analysis, embeddings for semantic search, vision for image understanding, and Whisper for audio transcription.
export OPENAI_API_KEY=sk-...
Reader Agent - Content Analysis & Memory Extraction
The Reader agent uses GPT-4o-mini to extract atomic memories from content, suggest tags, generate summaries, and classify relationships between memories.
{
"memories": [
{
"content": "User prefers PostgreSQL for JSONB support",
"confidence": 0.95,
"memory_type": "preference",
"entities": {
"technologies": ["PostgreSQL", "JSONB"]
}
}
],
"token_usage": {"input": 156, "output": 89},
"cost_cents": 0.0077
}
Searcher Agent - Semantic Search
The Searcher agent uses text-embedding-3-small to generate 1536-dimensional vectors for semantic search. Find content by meaning, not just keywords.
When you search for "authentication problems", the Searcher finds entries about "login issues", "JWT errors", and "session expiry" because the vector representations capture semantic similarity, not just keyword matches.
Asset Enrichment Agent - Image Understanding
The Asset Enrichment agent uses GPT-4o to analyze images, generate descriptions, and extract text through OCR. Make your image library searchable.
Generates detailed descriptions including:
Extracts text from:
Asset Enrichment Agent - Audio Transcription
Whisper transcribes audio files with automatic language detection and segment-level timing. Perfect for podcasts, voice memos, and meeting recordings.
{
"text": "Full transcription...",
"language": "en",
"duration": 145.5,
"segments": [
{
"id": 0,
"start": 0.0,
"end": 4.2,
"text": "Hello and welcome..."
}
]
}
Pay-as-you-go pricing. All costs tracked per request.
| Service | Model | Input Cost | Output Cost |
|---|---|---|---|
| GPT (Reader) | gpt-4o-mini | $0.15 / 1M tokens | $0.60 / 1M tokens |
| Embeddings (Searcher) | text-embedding-3-small | $0.02 / 1M tokens | - |
| Vision | gpt-4o-mini | ~$0.005-0.02 / image | |
| Whisper | whisper-1 | $0.006 / minute | |
Onelist tracks token usage and calculates costs for every API call. Monitor your spending in real-time through the dashboard or API responses.
Create an account at platform.openai.com and generate an API key.
Self-host Onelist with Docker and connect your OpenAI API key.