
Zero-dependency LLM inference engine. Fastest cold start, smallest VRAM footprint. CUDA + ROCm + Metal.

It was an AI-generated video of the goddess Lakshmi that convinced Eros Innovation founder Kishore Lulla that India needed a different approach to AI. “I was playing with some LLMs and wanted to make a video out of our goddess Lakshmi,” says Lulla, sitting d…

Official Python SDK for the Astrolinkers API — natal charts, talent profiles, Vedic calculations and engine-grounded LLM interpretations.

NeuGenM Launches First-Ever LLM Advertising in India, South Asia & Southeast Asia Through Exclusive Thrad Network Partnership

Structured LLM responses across OpenAI, Anthropic, DeepSeek, and Mistral (async, Pydantic schemas).