b7976
A popular AI tool rolls back its latest update, removing key model integrations.
Deep Dive
The developers of the llama.cpp project, a key tool for running AI models locally, have reverted a commit that added support for Qwen3.5 models. This action temporarily removes compatibility with these dense and mixture-of-experts AI models from the software. The reason for the rollback is not specified in the release notes. The project continues to provide pre-built binaries for major operating systems including Windows, macOS, Linux, and iOS.
Why It Matters
This affects developers and users who rely on this software to run the latest AI models efficiently on their own computers.