Models & Labs
OpenAI Enhances Voice Agents with Real-Time Models
The Rundown AIhigh confidence
Why it matters
- →OpenAI's new models significantly improve real-time reasoning in voice AI.
- →The models enable more natural and efficient voice interactions, reducing interruptions.
- →This advancement could shift the focus from text-based to voice-based AI systems.
OpenAI has launched three new real-time voice models, enhancing the capabilities of AI voice agents. The models, GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper, offer improved reasoning, tool use, and conversational flow. GPT-Realtime-2, in particular, shows a significant improvement in real-time reasoning, scoring 96.6% on Big Bench Audio. Companies like Zillow and Priceline are already integrating these models into their services, indicating a shift towards more natural and efficient voice interactions. This development suggests a move away from turn-based voice AI systems.
Read originalMore from The Rundown AI
More in Models & Labs
Models & Labsmodels
Llama.cpp b9075 Release Enhances CUDA Snake Activation
The b9075 release of llama.cpp brings a notable improvement for CUDA users by integrating the snake activation function into a single elementwise kernel. This enhancement is particularly advantageous for audio decoders like BigVGAN and Vocos, which previously depended on a more complex five-operation sequence. By streamlining these operations, the update promises better performance and efficiency across data types such as F32, F16, and BF16. This development reflects llama.cpp's ongoing focus on refining its CUDA capabilities, making it a more compelling option for developers dealing with complex activation functions.
llama.cpp Releases
Models & Labsmodels
Llama.cpp b9076 Release Expands Platform Support
The latest b9076 release of llama.cpp quietly expands its platform support, making it more versatile for developers across various systems. Notably, it now exposes child model information from the router's /v1/models endpoint, enhancing transparency and control for users. The update includes support for macOS Apple Silicon with KleidiAI enabled, as well as expanded compatibility with Ubuntu and Windows systems, including Vulkan and ROCm 7.2. This release doesn't introduce new models but strengthens llama.cpp's position as a flexible inference runtime across diverse hardware configurations.