llm-app-stack Releases

This release focuses on refactoring application settings, adding a Trophy System backend, and improving performance logging for ingredient sync operations. It also includes numerous bug fixes and dependency updates.

Jan 18, 2026

Introduces new probes and generators, improves documentation, and includes architectural updates to remove deprecated prompt initialization methods.

Dec 12, 2025

Nov 9, 2025

This release focuses on enhancing the Voice Activity Detection (VAD) model's robustness by reworking the training paradigm and improving stability on out-of-distribution data.

Nov 6, 2025

A maintenance release focused on PyTorch 2.9 compatibility, code quality improvements, and bug fixes in tuning utilities.

Nov 5, 2025

Silero VAD v6.0 introduces a new training algorithm for higher robustness, achieves 16% fewer errors on noisy data and 11% fewer errors on multi-domain validation, and adds C++ source support based on libtorch.

Aug 26, 2025

Minimal release notes: "## Changelog * d607566 chore: Update version to v1.4.1 in code and documentation"

Aug 24, 2025

Minimal release notes: "## Changelog * 6093f5a chore: remove release notes file and clean up version display * caa58a2 feat: Add 4 new API integ"

Aug 24, 2025

This security tool release adds custom wordlists, progress reporting, resource management, and security tool integrations including OWASP ZAP, Burp Suite, and Nessus, along with waybackurls integration.

Aug 24, 2025

D2 v0.7.1 adds ASCII rendering for diagrams, new arrow shapes including cross, underline support for class fields/methods, and customizable monospace fonts, with a breaking change affecting sketch font rendering.

Aug 19, 2025

Ultravox v0.6 improves language understanding, particularly for Hindi, adds noise token handling for non-human audio, and introduces new gemma3 and qwen3 model variants alongside VoiceBench evaluation support.

Aug 18, 2025

This release changes the default model to GPT-4.1-mini, which brings considerable quality improvements but may exhibit different behavior compared to the previous GPT-4o-mini default.

Jul 31, 2025

This security payload repository update introduces new vulnerability pages covering PHP extract() vulnerabilities and reverse proxy misconfigurations, along with enhanced exploitation techniques for command injection, CSV injection, file inclusion, and SQL injection.

Jul 26, 2025

TinyTroupe alpha release introduces the TinyTroupe paper as a preprint, adds action correction mechanisms for TinyPersons, enables parallel agent generation and simulation, and introduces the InPlaceExperimentRunner for A/B testing.

Jul 16, 2025

May 11, 2025

May 11, 2025

This workout management application update adds reverse proxy support, ingredient synchronization, full text search, and Prometheus metrics export, along with numerous Docker image improvements and flexible routines.

Apr 5, 2025

Mar 9, 2025

PaddleSpeech r1.5.0 upgrades from Paddle 2.5.1 to Paddle 3.0.0-beta, adds AudioTools toolkit support for DAC training/inference, and introduces GCU Backend support with NPU and MLU adaptations for SpeedySpeech.

Mar 5, 2025

Released Ultravox v0.5 with significant performance improvements, including 60% better transcription accuracy, 18% better speech QA, and support for 42 languages.

Feb 11, 2025