LLaMA-Omni Releases
112 releases of ictnlp/LLaMA-Omni
- v2.7
This release addresses tool routing stability and introduces new capabilities including a Cerebras model provider and browser proxy support.
Jul 27, 2026
- shadcn@4.15.0
This release introduces a new public API for programmatically installing registry items, allowing developers to add components to their projects without using the command-line interface.
Jul 25, 2026
- v2.6
This release focuses on stability and performance improvements for the Agent Zero platform, consolidating the goal tool architecture and enhancing secret handling.
Jul 23, 2026
- shadcn@4.14.1
This patch update integrates Base UI Toast support into the shadcn library, enhancing the notification system for user interfaces.
Jul 23, 2026
- shadcn@4.14.0
This minor release adds functionality to assist users with migrating icons within their projects, streamlining the process of updating icon assets.
Jul 22, 2026
- v152.0.4-beta.28
This beta release focuses on fixing mouse trajectory logic and edge bounds in the 'juggler' module to ensure compatibility with Firefox 146.
Jul 19, 2026
- v2.5
A major update featuring a global model presets overhaul, new default models, LaTeX rendering in chat, and enhanced desktop session integration.
Jul 17, 2026
- shadcn@4.13.1
A patch release prioritizing security fixes to prevent credential leakage and path traversal, alongside the addition of React Aria support.
Jul 17, 2026
- v152.0.4-beta.27
Marked as the new latest stable release to replace outdated FF135 builds, addressing page hang issues and WebRTC leaks.
Jul 16, 2026
- 4.18.0
A feature update for Cloudreve introducing one-click recycling, sidebar pinning for Pro users, and various security and usability fixes.
Jul 15, 2026
- v0.1.1
A minor feature update for the socid-extractor adding support for social media platforms and academic schemes, along with fixes for existing extractors.
Jul 13, 2026
- v2.4
A comprehensive update introducing crash-safe chat persistence, new AI provider integrations (NVIDIA NIM, Gemini CLI), and significant improvements to terminal handling, preset management, and UI controls.
Jul 10, 2026
- v2.3
Focuses on browser streaming capabilities, introduces three new built-in plugins (Orchestrator, Slash Commands, Goal Management), and migrates several providers to chat-completions transport.
Jul 9, 2026
- @shadcn/react@0.2.1
This is a patch release for the shadcn/ui React component library focusing on critical bug fixes for the MessageScroller component to improve stability during streaming content.
Jul 8, 2026
- v0.5.20feat: thinking level picker + suffix support across providers version 0.5.20
This release introduces a per-model thinking level picker that appends a suffix to model names for forced reasoning effort across providers. It includes various fixes for reasoning content handling, token counting, and new Farsi language support.
Jul 7, 2026
- v0.5.18v0.5.18: cached token tracking, provider fixes version 0.5.18
This release focuses on improving usage tracking by adding cached token calculation and fixes for various providers including Claude, Kiro, and Kimchi. It also introduces new models and capabilities for NVIDIA and ClinePass.
Jul 3, 2026
- v2.2
A major update improving onboarding UX with a deferred model setup gate and unified provider selection. Enhances the chat composer with Markdown code blocks and stabilizes LiteLLM provider fallback.
Jul 2, 2026
- v0.9.0
This is a major release introducing streaming model support through the integration of transcribe.cpp. It also expands model support and includes various bug fixes, documentation updates, and performance improvements.
Jul 1, 2026
- v1.90.0
This release focuses on Docker image security verification using cosign and introduces support for the Claude Fable 5 model across multiple providers. It also includes significant UI improvements and a refactor of the Bedrock toolSpec implementation.
Jun 27, 2026
- v0.5.12feat: token-saver dashboard, Venice AI, Blackbox catalog overhaul version 0.5.12
This release adds a token-saver dashboard, introduces the Venice AI provider, and overhauls the Blackbox provider catalog.
Jun 26, 2026