AI that answers from verified data, not training data.
Bring your own LLM API key. Aplicious fetches live data, structures it as context, and grounds the model — so every response is anchored to real numbers. You pay your LLM provider directly.
Supported LLM providers
| Provider | Models | Status |
|---|---|---|
| OpenAI | GPT-4o, GPT-4, GPT-3.5 Turbo | LIVE |
| Anthropic | Claude 3.5 Sonnet, Claude 3 Opus, Haiku | LIVE |
| Google Gemini | Gemini 1.5 Pro, Gemini 1.5 Flash | LIVE |
| Mistral | Mistral Large, Mistral 7B | LIVE |
| Groq | Llama 3, Mixtral (fast inference) | LIVE |
OCR, Transcription, and TTS require an OpenAI key — they use GPT-4o Vision and Whisper. All text AI endpoints work with any listed provider.
Intelligence endpoints
Conversational AI with live Aplicious data as context. One chat config, unlimited users each with their own session history.
Single-shot AI summaries. One narrative template — run it for every customer and get a personalised data-grounded response.
Vedic astrology interpretations grounded in Swiss Ephemeris data. Chart narratives, dasha interpretation, transit forecasts.
AI-powered media & text endpoints
Extract text from images and PDFs
OpenAI (GPT-4o Vision)
Audio to text in 50+ languages
OpenAI (Whisper)
Natural voice synthesis, 6 voices
OpenAI (TTS)
Text summarisation in 3 formats
Any provider
100+ language pairs
Any provider
Document and sentence-level sentiment
Any provider
Anti-hallucination guardrails
Every AI call through Aplicious includes a grounding envelope in the system prompt. The model is instructed to distinguish between VERIFIED (from live data), DERIVED (interpretation), and GENERAL (model knowledge) claims. Numeric claims are validated before the response is returned.
Every response includes an _audit block showing exactly what verified data was supplied — so you can inspect what the model had access to.
Start building AI products today
Connect your LLM key from the dashboard and make your first AI call in minutes.
Connect LLM key →