Lossless AI model compression - ~34% smaller with bit-identical weights; the autopilot profiles your machine, picks the highest fidelity that runs, and streams models bigger than your RAM.
World's most deployable time series foundation model — 200K-6.5M params, zero-shot forecasting, streaming RNN inference, ONNX edge deployment, runs on Raspberry Pi
A complete inference runtime for open-weight large language models, enabling efficient execution through streaming weights, quantization, and memory-aware scheduling
