Open Source
Tencent Releases Hy3: Open MoE LLM
Lev Selectorhigh confidence
Why it matters
- →Supports open-source AI development.
- →Provides a powerful tool for researchers.
- →Enhances language model performance.
Tencent has introduced Hy3, an open-source mixture of experts (MoE) large language model. This release is part of Tencent's efforts to contribute to the open-source AI community, providing a powerful tool for developers and researchers. Hy3 is designed to offer enhanced performance through its MoE architecture, which allows for more efficient processing of language tasks.
Read originalMore from Lev Selector
© Lev SelectorModels & Labsmodels
New AI Models Released: Grok 4.5, GPT-5.6, Gemini 3.5
Several new AI models have been released, including Grok 4.5, GPT-5.6, and Gemini 3.5.
Lev Selector
© Lev SelectorModels & Labsmodels
OpenAI Optimizations Halve Inference Costs
OpenAI has implemented optimizations that reduce inference costs by 50%.
Lev Selector
© Lev SelectorAgentsagents
Composio SDK Connects AI Agents to Hundreds of Apps
The Composio SDK now allows AI agents to connect with hundreds of applications.
Lev Selector
More in Open Source
Open Sourcemodels
llama.cpp b9947 Release Expands Platform Support
The latest b9947 release of llama.cpp continues its trend of broadening platform compatibility, though without major new features. Notably, the release includes support for ROCm 7.2 on Ubuntu x64, which is significant for AMD GPU users seeking alternatives to NVIDIA's CUDA. While KleidiAI support for Apple Silicon remains disabled, the release still covers a wide array of systems, from Windows CUDA 13 to Ubuntu Vulkan. This update solidifies llama.cpp's role as a versatile inference runtime, though it doesn't introduce groundbreaking changes.
llama.cpp Releases
Open Sourcemodels
llama.cpp b9949 Release Expands Platform Support
The latest b9949 release of llama.cpp continues its trend of broadening platform compatibility, notably adding support for ROCm 7.2 on Ubuntu x64, which is a significant step for AMD GPU users. This release also includes updates for Windows with CUDA 12 and 13, enhancing its utility for developers working across different hardware configurations. While KleidiAI support for macOS Apple Silicon is disabled, the release still marks a steady expansion of llama.cpp's reach across diverse systems. This update doesn't introduce new models but strengthens the framework's versatility and accessibility for developers.
llama.cpp Releases
Open Sourcemodels
llama.cpp b9950 release focuses on platform fixes
The b9950 release of llama.cpp is a technical update that addresses specific platform issues and enhances code reliability with new unit tests for llama-batch. It resolves build problems on Win32 and introduces assertions for methods that are not yet implemented. While this update doesn't bring new models or groundbreaking features, it ensures compatibility across a wide array of systems, including macOS, Linux, Windows, and openEuler. This release is a step towards refining the software's robustness and usability across different hardware configurations, making it more stable and reliable for developers.
llama.cpp Releases