Topic: macos
17 stories found
Sunday, November 1, 2026
ggml/llama.cpp releases: b12032
The ggml/llama.cpp project released version b12032, which includes a change to try using PkgConfig for LevelZero includes and libraries (#31076). This update is important as it may improve compatibility on macOS Apple Silicon (arm64) systems.
Saturday, October 31, 2026
ggml/llama.cpp releases: b12024
The ggml/llama.cpp project updated its codebase to reshape tensors from 3D to 2D for language model fine-tuning, addressing issue #30831. This change is significant as it optimizes memory usage and potentially improves performance in machine learning tasks.
ggml/llama.cpp releases: b12018
The ggml/llama.cpp project released version b12018, which includes a fix for f32 to q2_0/q4_0/q5_0 rounding in CPY; this update is important for improving the precision of quantization processes.
Vercel Labs Open-Sources agent-desktop for Agent Computer Use
Vercel Labs open-sourced agent-desktop, an Apache-2.0 Rust CLI enabling AI agents to interact with computers across macOS, Windows, and Linux by reading accessibility trees. This tool aims to improve AI integration and usability on different operating systems without relying on pixel-based interactions, marking a step forward in accessible AI technology.
Friday, October 30, 2026
ggml/llama.cpp releases: b12017
The ggml/llama.cpp project removed the fs_path_to_utf8() function in version b12017, as detailed in issue #31424 and signed off by Adrien Gallouët. This update is part of an ongoing effort to improve the library, though its specific impact on functionality or performance isn't detailed here.
ggml/llama.cpp releases: b11998
The ggml/llama.cpp project has released an update that includes extending DSA and sdpark for the mistral le chonk model and fixing CI issues. This update is significant as it enhances model functionality and improves development processes, making the project more robust and user-friendly.
Monday, October 26, 2026
ggml/llama.cpp releases: b11910
The ggml/llama.cpp project released a new version that includes an update to write KV-cache views in set_tensor, preventing state restores from being dropped. This update is significant as it enhances the stability and functionality of the model, particularly useful for applications requiring continuous inference sessions.
Sunday, October 25, 2026
ggml/llama.cpp releases: b11887
The ggml/llama.cpp project released update b11887, which includes adjusting the default threads for test-llama-context. This update is important as it optimizes multi-threading tests, potentially improving performance and reliability in handling large language model contexts.
Saturday, October 24, 2026
Ollama releases: v0.40.11
Ollama v0.40.11 adds speech functionality on macOS, allowing text-to-speech generation during model runs and supporting audio streaming via the API, enhancing accessibility and user experience. This update is significant as it broadens Ollama's capabilities, making it more versatile for various applications.
Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for AI Agents, No Containers
Anthropic has released sandbox-runtime (srt), an open-source OS-level sandbox for restricting access of AI agents and tools on macOS and Linux without using containers. This innovation aims to enhance security by directly managing permissions at the operating system level, crucial for safeguarding against potential risks posed by AI systems.
🌿 That's all for now. Come back tomorrow.
17 of 17 items shown. Sources: 169 days indexed.