Models & Labs
GitHub API Adds Team-Level Copilot Metrics
GitHub Changeloghigh confidence
Why it matters
- →Provides granular insights into Copilot adoption and usage at the team level.
- →Enables targeted enablement and investment strategies based on team performance.
- →Facilitates detailed analysis of Copilot's impact across different development environments.
GitHub has introduced team-level Copilot usage metrics through its API, allowing enterprise administrators and organization owners to track Copilot adoption and activity by team. The new feature provides detailed insights into active users, completions, and other metrics, broken down by language, IDE, feature, and model. This development enables organizations to identify which teams are leading in adoption and which may need further support. The metrics are accessible via REST API endpoints, though no dashboard is available for these team-level insights.
Read originalMore from GitHub Changelog
© GitHub ChangelogCoding Toolscoding
GitHub Copilot App Enters Technical Preview
GitHub has launched a technical preview of its Copilot app, offering a desktop experience for agentic development directly within GitHub. This app allows developers to start sessions from existing work artifacts like issues or pull requests, maintaining context and isolation across tasks. It supports automation of routine tasks and integrates seamlessly with GitHub's existing review and merge processes. This development marks a step towards more integrated and efficient workflows for developers using GitHub's ecosystem.
GitHub Changelog
© GitHub ChangelogMore in Models & Labs
Models & Labsmodels
llama.cpp b9145 release addresses SYCL memory issues
The latest llama.cpp release, b9145, tackles a significant issue with SYCL's memory allocation on multi-GPU systems, particularly those using Intel Arc Pro GPUs. By replacing sycl::malloc_device with zeMemAllocDevice, the update drastically reduces system RAM usage from 60 GiB to just 6.7 GiB for a 15.6 GiB model, preventing out-of-memory crashes without sacrificing performance. This change is crucial for developers working with large models on multi-GPU setups, as it ensures more efficient memory management. The update also includes several improvements and bug fixes, enhancing the robustness of the SYCL backend.
llama.cpp Releases
Models & Labsmodels
Llama.cpp Adds Qwen3.5 Tokenizer Handler
Llama.cpp's latest release enhances its capabilities with a non-backtracking tokenizer handler specifically designed for Qwen3.5. This update significantly improves Unicode tokenization, addressing stack overflow issues that occur with long inputs. By adapting the previous Qwen2 fix to meet Qwen3.5's regex requirements, including support for accent marks, the update ensures more reliable text processing. Developers can now expect more stable performance when handling complex Unicode inputs, benefiting from the robust tokenization across different operating systems and hardware configurations. This means smoother operations on platforms like macOS with KleidiAI, Ubuntu with ROCm 7.2, and Windows with CUDA 12 and 13.