v2.4.1
facebookresearch/tribev2v2.4.1Aug 24, 2026by claxman
Upgrading?
AI Summary
A follow-up to the 2.4.0 release enhancing the 'Ask' feature with streaming responses and persistent chats, while adding Microsoft Teams 1:1 call detection.
Key Highlights
- Streaming answers for Ask feature
- Persistent chats support (restored on reopen)
- Microsoft Teams 1:1 call detection
- macOS in-app updates (consistent with Windows)
New Features
- Streaming answers
- Persistent chats
- Microsoft Teams 1:1 call detection
- macOS in-app updates
Full Release Notes
Follow-up improvements to the 2.4.0 meeting-intelligence release. ## Ask - **Streaming answers** — Ask now types out token-by-token, in both per-meeting Ask and "Ask my meetings", so you don't wait for the whole response. - **Persistent chats** — per-session Ask is restored when you reopen a meeting, and "Ask my meetings" now supports multiple saved chats (new / switch / rename / delete, auto-titled from your first question). A turn is no longer lost if you navigate away mid-answer. - **Input pinned to the bottom** like a normal chat, with messages scrolling above it. ## Meeting detection - **Microsoft Teams 1:1 calls** are now detected, not just scheduled meetings. ## Updates (macOS) - macOS now updates **in-app like Windows**: click Update, it downloads, then "Restart & update" installs — no more manual disk-image swap. (Existing 2.4.0 Mac users receive 2.4.1 via the installer one last time; in-app self-update applies from 2.4.1 onward.) ## Downloads - **macOS 12+ (Apple Silicon):** [Raven-Mac-2.4.1-Installer.dmg](https://github.com/Laxcorp-Research/project-raven/releases/download/v2.4.1/Raven-Mac-2.4.1-Installer.dmg) (Developer ID signed + notarized) - **Windows 10/11 (x64):** [Raven-Windows-2.4.1-Setup.exe](https://github.com/Laxcorp-Research/project-raven/releases/download/v2.4.1/Raven-Windows-2.4.1-Setup.exe) Enter your own API keys in Settings on first launch. Mac artifacts are built, signed, and notarized by CI and attached automatically.
Related Repositories
rasbt/LLMs-from-scratch★ 107,042
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
lyogavin/airllm★ 40,153
AirLLM 2.8T LLM inference with single 4GB GPU
NirDiamant/RAG_Techniques★ 29,958
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. Each technique has a detailed notebook tutorial.
microsoft/OmniParser★ 25,539
A simple screen parsing tool towards pure vision based GUI agent
karpathy/nn-zero-to-hero★ 24,906
Neural Networks: Zero to Hero