laya-mlx Releases

This release introduces support for Qwen3.5 and Gemma 4 models, alongside performance optimizations like reduced VRAM usage and improved reproducibility. It also adds an integrated benchmarking system and fixes various multi-GPU reporting and startup speed issues.

May 5, 2026

Minimal release notes: "**Full Changelog**: https://github.com/mukul975/cve-mcp-server/commits/v0.1.0"

Apr 14, 2026

This release adds GPU device control, new recognizers for the UK and US, and ONNX Runtime support for GLiNER, alongside critical security fixes for CVE vulnerabilities and supply chain hardening.

Mar 18, 2026

This release enhances network connectivity options by adding support for HTTP proxies and custom CA certificates to the configuration system.

Feb 20, 2026

This release adds a new LoRA-based abliteration engine with 4-bit quantization, broad support for Vision-Language (VL) models, and the ability to save and resume optimization progress. It also includes memory management improvements and various stability fixes.

Feb 14, 2026

This release enhances the platform with Language models integration (LangExtract), adds recognizers for various countries including Thailand and Korea, and optimizes Docker images with Python 3.13 support and HEALTHCHECKs.

Feb 12, 2026

This release introduces support for the new Voyage 4 family of embedding models, featuring high performance rankings and advanced capabilities like long context windows.

Jan 16, 2026

Minimal release notes: "**Full Changelog**: https://github.com/chonkie-inc/catsu/compare/v0.1.5...v0.1.6"

Jan 12, 2026

Minimal release notes: "**Full Changelog**: https://github.com/chonkie-inc/catsu/compare/v0.1.4...v0.1.5"

Jan 7, 2026

Minimal release notes: "**Full Changelog**: https://github.com/chonkie-inc/catsu/compare/v0.1.3...v0.1.4"

Jan 7, 2026

Minimal release notes: "- update the inference pipeline and models to support unlimite audio length - update the models cards"

Dec 22, 2025

This release addresses multiple critical bug fixes including permission denial errors after Android upgrades, state restoration issues, and startup problems on Meizu devices. It also improves build script handling and fixes message parsing.

Dec 17, 2025

Dec 12, 2025

A critical update to resolve immediate failure upon upgrading to Android 16 beta, alongside fixes for frame memory leaks on Windows and improvements to virtual display presentation.

Sep 27, 2025

This release introduces the Korean Resident Registration Number (RRN) recognizer and integrates Azure Health Data Services (AHDS) for de-identification, alongside significant CI/CD improvements like OIDC publishing and multi-platform Docker support.

Sep 9, 2025

Implements internationalization (i18n) support and adds Llama API integration. Synchronizes with browser-use v0.5.11 and removes the validator agent in favor of the planner.

Aug 25, 2025

Optimizes the Ai Workflow to accept images, audio, video, and file types as input parameters. Introduces a new AP Workflow for improved quick access.

Aug 11, 2025

Introduces a new AI Workflow powered by AI Power that supports mustcache syntax. Adds Portuguese-BR language support.

Jul 23, 2025

Improves Google Sheets integration with stricter field validation and migrates the Google Sheet Picker component to the ChromeV3 specification. Fixes document link accessibility.

Jul 23, 2025

Focuses on caching improvements, resolver workflow enhancements, and charset normalization updates to support multi-language content better.

May 21, 2025