Abhishek Yadav

🚀 Big News for Developers! 🚀

Google just launched Gemini Code Assist a free AI-powered coding assistant! 🎉

✅ Highest free usage limits up to 180,000 code completions/m
✅ AI-powered code reviews for GitHub (public & private repos)
✅ Runs in VS Code & JetBrains IDEs

Try it out 🔗👇
bit.ly/4id2lDi

जख्म बाते नही
उसके भाव देते है

CodeTree 🌳🌳🌳 :
The first unified framework combining tree-based strategy exploration + execution feedback + LLM agent guidance for code generation.

🖇️ Paper: arxiv.org/pdf/2411.04329

📈 Setting new standards with GPT-4:

95.1% HumanEval
98.7% MBPP
43.0% CodeContests

#pdf is an important file format. But why do you need Acrobat from the evil Adobe while you can edit pdf documents with #LibreOffice. LibreOffice draw is well suited for editing and creating pdf documents. 
Let the #foss be with you. #floss #opensource

Implement 4 agentic patterns
for your next AI project 🤖

Reflection 💭
Tool 🛠️
Planning 📋
Multiagent 👥

github.com/neural-maze/...

aisuite : Easily use LLMs from multiple providers in one app! 🛠️

it often tedious task to integrate multiple providers when building apps.

Aisuite simplifies this by letting you switch between models with a single string like:

Open-source repo: github.com/andrewyng/aisu…

Some nice collection of resources for LLMs.

llmresourceshub.vercel.app

🍓Marco-o1! Large Reasoning Models for Open-Ended Solutions.

🎯 Built and released a CoT dataset to activate LLMs' reasoning abilities.

💡 Integrated LLMs with MCTS to expand the solution space.

🔬 Exploited action granularities in MCTS and introduced a reflection mechanism.

AWS releases Multi-Agent Orchestrator!

A flexible framework for managing multiple AI agents and handling complex conversations.

Features include:
- Dynamic query routing
- Python & Typescript support
- Streaming support
- Context management
- Run locally or on any cloud platform

Studio by hcompany the most advanced agent to date — designed to navigate web interfaces through pixel-level interpretation and semantic understanding. You can now turn your instructions into action with human-like precision.
www.hcompany.ai

DeepSeek-R1-Lite-Preview is now live:

🔍 Achieves o1-preview-level performance on AIME & MATH benchmarks.

💡 Transparent thought process in real-time.

🛠️ Open-source models & API coming soon!

#AI #MachineLearning #DeepSeek
chat.deepseek.com

What effect you would like say ??
#bluesky #twitter #thread

LLaVA-o1 is the first visual language model capable of spontaneous, systematic reasoning, similar to GPT-o1!

🤯 11B model outperforms Gemini-1.5-pro,GPT-4o-mini, and Llama-3.2-90B-Vision-Instruct on six multimodal benchmarks.

AI agent orchestration framework for building LLM apps with RAG and multiple AI Agents in Python.

100% open source.

Sure

Chinese startup 01 .ai trains competitive LLM using 95% fewer resources through innovative engineering optimization.

GitHub - codecrafters-io/build-your-own-x: Master programming by recreating your favorite technologies from scratch. Master programming by recreating your favorite technologies from scratch. - codecrafters-io/build-your-own-x

github.com/codecrafters...

You can literally build whatever you want

🧬Evo, the first foundation model trained at scale on DNA, is a Rosetta Stone for biology.

DNA, RNA, and proteins are the fundamental molecules of life—and cracking the code of their complex language is an ongoing grand challenge.

www.science.org/doi/10.1126/...

Starting my migration process from X