Models & Labs
BreezeTTS2: Real-time local voice synthesis
Sam Witteveenhigh confidence
Why it matters
- →Enables private, real-time TTS for interactive apps without cloud dependency.
- →Open weights allow full customization and local deployment of a 3B parameter model.
- →Low latency performance makes it viable for voice agents and live synthesis tasks.
BreezeBlue has released BreezeTTS2, a 3-billion parameter open-weight text-to-speech model designed for local, real-time inference. The model supports voice cloning, voice direction, and multi-lingual generation while maintaining low latency suitable for interactive applications. It is available on Hugging Face under an open license, allowing developers to deploy it entirely on-premise or on edge devices without relying on cloud APIs.
Read originalMore from Sam Witteveen
More in Models & Labs
Models & Labscoding
llama.cpp b10996 adds CUDA 13 and ROCm 10.0 builds
This release quietly closes the hardware gap for local inference by adding native support for CUDA 13 and ROCm 10.0 across Linux and Windows. NVIDIA users can now leverage newer driver stacks without waiting for upstream updates, while AMD GPU owners finally get first-class ROCm 10 binaries that match the maturity of their CUDA counterparts. Apple Silicon builds remain available but KleidiAI is explicitly disabled here, suggesting a focus on stability over new kernel optimizations for this specific iteration. The inclusion of openEuler support for Huawei's Ascend chips further broadens the ecosystem beyond standard x86 and ARM consumer hardware.
llama.cpp Releases
Models & Labsmodels
Llama.cpp b10981 release enhances OpenVINO support
The latest b10981 release of llama.cpp brings significant improvements to OpenVINO integration, particularly in optimizing stateful decode and GPU MoE inference. By addressing issues like stateful decode errors and enhancing the handling of sliding-window layers, this update ensures more reliable and efficient model performance. The release also introduces new features such as the GGML_OPENVINO_REQUANT_KQUANT for 4-bit requantization and improved handling of multi-head models. These changes make llama.cpp more robust and versatile, especially for developers working with complex AI models on diverse hardware setups.