awesome-virtual-try-on Releases

Bug fixes for authentication caching, LDAPv3 version checks, AMR consistency, and the addition of configurable rate limits for OIDC.

Apr 9, 2026

Focuses on UI enhancements, PDF viewer improvements, and new integrations such as Polars dataframes and Portuguese translation.

Apr 7, 2026

A comprehensive update focusing on performance improvements across CUDA and Metal backends, new model support including Gemma 4 and Qwen 3.5, and enhanced quantization capabilities.

Apr 2, 2026

A minor patch release that cleans up shell aliases during initialization to prevent conflicts.

Mar 31, 2026

This patch release focuses on security improvements and infrastructure updates. It migrates the npm publishing mechanism to use OIDC and fixes a critical vulnerability regarding session ownership during WebSocket restores.

Mar 27, 2026

This release addresses several security vulnerabilities and bugs, particularly around authentication, OIDC integration, and UI stability. It includes fixes for excessive filter escaping, language cookie sanitization, and a stuck state issue during 2FA enrollment.

Mar 14, 2026

A feature-rich release introducing new configuration options, UI enhancements, and internationalization support. Highlights include configurable settings for cookies and avatars, improved message management features, and support for new languages.

Mar 5, 2026

Adds an MCP bundle for easy installation and enhances the loop command with verbose output and metadata support for tasks.

Feb 4, 2026

A major update introducing a new CLI, Prefix Caching for performance, extensive model support (GLM-4, Gemma 3n, Qwen 3), and dynamic model loading capabilities.

Jan 28, 2026

A minor UI-focused release for Chainlit adding new input widgets and settings toggles to improve user experience.

Jan 20, 2026

Improves task filtering and concurrency safety with new filters, cross-process file locking, and updated Codex CLI models.

Jan 15, 2026

Introduces an automated loop command for task execution with presets and fixes Zod schema validation issues.

Jan 11, 2026

Focuses on performance improvements with a 2x faster CLI startup and smarter project root detection.

Dec 28, 2025

This release introduces enhanced logging capabilities and expands SSL configuration options to all editions. It addresses a wide range of stability issues including certificate renewals, proxy IP display, and custom rule management. Additionally, the Anti-Bot Challenge service has been updated to improve overall system reliability.

Nov 5, 2025

This release introduces major model support including Llama 4 and Qwen 3, alongside a revamped Web chat app and expanded multimodal capabilities. Significant performance optimizations were made to Metal and CUDA backends, including advanced quantization methods and improved attention schedulers.

Jun 10, 2025

Minimal release notes: "**Full Changelog**: https://github.com/agentheroes/agentheroes/commits/v0.0.1"

Apr 8, 2025

v0.5.0 adds support for several new models like Gemma 3 and Qwen 2.5 VL, while introducing native tool calling and Tensor Parallelism capabilities. The release focuses heavily on performance, featuring a 30x speedup in ISQ times on Metal and integrating FlashAttention V3 into the PagedAttention system.

Mar 24, 2025

This release focuses on expanding model support with DeepSeek V2, V3, and R1, alongside the introduction of Imatrix quantization and BNB quantization. It also adds support for FP8 dequantization on both CUDA and Metal, integrates the llguidance library for constraints, and brings Metal PagedAttention support.

Jan 22, 2025

v0.1 serves as a placeholder release for model weights, referencing external storage due to size constraints, specifically for the MMAudio project.

Dec 7, 2024

Minimal release notes: "- Added `autoawq` model family - Published multiple docker images supporting multiple BLAS backends (llama.cpp) - Remo"

Nov 13, 2023