Forwarded from Laylo
Laylo's Changelog
Formatting Improvements 🎨
· Enhanced markdown formatting options in responses for better readability and aesthetics.
· Improved overall visual experience with cleaner, more intuitive layouts.
Chat History Enhancements 🔄
· Updated chat history to reflect exactly what users see in their messenger interfaces. No more discrepancies between my memory and the UI.
Tool Optimization 💡
· Reworked internal processes for handling web searches and other tools, making them faster and more reliable.
· Now provide real-time feedback on tool usage, giving you full visibility into when and how specific functions are being utilized.
Hidden Replies Support 🔒
· Introduced support for hidden replies. Messages starting with
Quote Reply Functionality 📝
· Implemented quote reply functionality, allowing Laylo to see exactly which part of a message is being replied to when quoting text.
---
For more information on the sources and instructions to run your own instance, visit the GitHub repository for tg-ollama.
Formatting Improvements 🎨
· Enhanced markdown formatting options in responses for better readability and aesthetics.
· Improved overall visual experience with cleaner, more intuitive layouts.
Chat History Enhancements 🔄
· Updated chat history to reflect exactly what users see in their messenger interfaces. No more discrepancies between my memory and the UI.
Tool Optimization 💡
· Reworked internal processes for handling web searches and other tools, making them faster and more reliable.
· Now provide real-time feedback on tool usage, giving you full visibility into when and how specific functions are being utilized.
Hidden Replies Support 🔒
· Introduced support for hidden replies. Messages starting with
// will be ignored by the system but still visible to users for note-taking or mentions of Laylo messages without engagement.Quote Reply Functionality 📝
· Implemented quote reply functionality, allowing Laylo to see exactly which part of a message is being replied to when quoting text.
---
For more information on the sources and instructions to run your own instance, visit the GitHub repository for tg-ollama.
😢1
Laylo
Laylo's Changelog Formatting Improvements 🎨 · Enhanced markdown formatting options in responses for better readability and aesthetics. · Improved overall visual experience with cleaner, more intuitive layouts. Chat History Enhancements 🔄 · Updated chat history…
Please open Telegram to view this post
VIEW IN TELEGRAM
🚀 Introducing Kimi k1.5 — an o1-level multi-modal model
- Sota short-CoT performance, outperforming GPT-4o and Claude Sonnet 3.5 on 📐AIME, 📐MATH-500, 💻 LiveCodeBench by a large margin (up to +550%)
- Long-CoT performance matches o1 across multiple modalities (👀MathVista, 📐AIME, 💻Codeforces, etc)
Apparently DeepSeek is not the only new Chinese gpt-o1-killer
Please open Telegram to view this post
VIEW IN TELEGRAM
👏1😢1
Please open Telegram to view this post
VIEW IN TELEGRAM
Forwarded from Laylo
Laylo AI: Exciting New Features and Updates!
In recent developments, Laylo AI has expanded its capabilities significantly. Here are some of the latest features that users can now enjoy:
· Image Search: Laylo can now help you find images on the web.
· NSFW Control: For a more personalized experience, Laylo now allows users to control NSFW content with the
· Custom Behavior: With the
Get ready for a more interactive and personalized experience with Laylo AI! Check out the source code on GitHub.
In recent developments, Laylo AI has expanded its capabilities significantly. Here are some of the latest features that users can now enjoy:
· Image Search: Laylo can now help you find images on the web.
· NSFW Control: For a more personalized experience, Laylo now allows users to control NSFW content with the
/nsfw on|off feature.· Custom Behavior: With the
/extremely <behavior> command (e.g., /extremely lazy), you can tailor Laylo's responses and behavior according to your preferences. Simply use /extremely without any argument to revert back to default settings.Get ready for a more interactive and personalized experience with Laylo AI! Check out the source code on GitHub.
😁1🤯1😢1
Fedora Project is gonna focus on AI in 2025 leveraging IBM Granite open models (no specific cases mentioned)
They are also hiring 2 devs to the desktop team
➡️ GNOME Blog
They are also hiring 2 devs to the desktop team
Please open Telegram to view this post
VIEW IN TELEGRAM
😢2⚡1👍1👏1
#TIL Whenever you copy any cells in LibreOffice Calc, it copies both data (text) and an image of a selection
🤔2
Media is too big
VIEW IN TELEGRAM
Nicer visuals, UI screens, extended API, resources and animations update; Client is now fully in TS, Godot part is restructured and rewritten for scalability
P.S. Godot is still super easy, I don't use any resources except docs and simple AI questions (well I've also checked 1 minute video on shaders)
Please open Telegram to view this post
VIEW IN TELEGRAM
🔥4👍2🤯1
- Added support for Twitter videos (
x.com/.../status/...)- Fixed video downloading (sometimes only audio was downloaded)
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
This media is not supported in your browser
VIEW IN TELEGRAM
Zed predictive edit (code completions) is now released for everyone (from beta) with open-source, open-dataset Zeta model, which is a Qwen2.5-Coder-7B fine-tuned by Zed developers
It can write new and edit existing code and jump between lines allowing it to modify relevant parts of the codebase regardless of cursor position
P.S. Freefor now & built-in in Zed
It can write new and edit existing code and jump between lines allowing it to modify relevant parts of the codebase regardless of cursor position
P.S. Free
🔥1🤔1
- Reworked prompting
- Streaming responses (doesn't wait for each message to be fully generated before responding)
- "typing" note
- Ability to respond with a message before using a tool, and edit it with an actual response afterwards
- Leverages structured outputs to constraint response format, making responses much more consistent and reliable
- Security measures (filesystem, network access via tools)
- Proper .env usage
-
/ai current preferences-
/ai [field] [value] to control preferences-
limit preference to enable display message usage limit- Deno rewrite
- More solutions, bigger README
Please open Telegram to view this post
VIEW IN TELEGRAM
🔥2
Please open Telegram to view this post
VIEW IN TELEGRAM
facebook.com/share/...)Please open Telegram to view this post
VIEW IN TELEGRAM