Models
timestral
mistralai/timestral
A family of time-series foundation models for multivariate probabilistic forecasting with lightweight inference code and leading zero-shot benchmark accuracy.
Related projects
VibeVoice54.9kMicrosoft's open-source frontier voice AI series providing streaming text-to-speech synthesis and recognition with a CPU-capable inference engine and demos for developers.
kronos41.5kReleases the Kronos family of foundation models for financial market candlestick sequences, including multi-scale pretrained weights and hierarchical discrete tokenizers for quantitative researchers and forecast developers.
ChatTTS39.9kAn open-source text-to-speech model for conversational dialogue with pretrained weights, multi-speaker support and fine-grained prosody control.
voxcpm39.5kAn open-source 2B tokenizer-free multilingual TTS model for voice design, controllable cloning, and studio-quality synthesis.
timesfm34.8kGoogle’s pretrained time-series foundation model for zero-shot multivariate forecasting with PyTorch and MLX inference code.
unlimited-ocr28.3kBaidu’s long-horizon document parsing OCR model with pretrained weights and Transformers, vLLM, and SGLang inference paths.