Latest Open Source Projects
deepclaude
deepclaude
TLDR: DeepClaude is a high - performance LLM inference API. It combines DeepSeek R1's reasoning and Anthropic Claude's creative and code generation capabilities, offering features like zero latency, security, high configurability, and is open - source. It allows users to use their own API keys and provides a unified interface to leverage the strengths of both models.
open-r1
open-r1
TLDR: This repo is a fully open reproduction of DeepSeek-R1. The goal is to build the missing parts of the R1 pipeline. It includes scripts for training and evaluating models, generating synthetic data, and has a Makefile for easy running of commands. The plan of attack involves replicating R1-Distill models, creating a pure RL pipeline, and showing multi-stage training. It supports training with DDP or DeepSpeed ZeRO-2 and ZeRO-3, and provides different methods for evaluating models depending on their size and hardware requirements.
open-operator
open-operator
TLDR: This is a proof of concept. It requires pnpm for installation. Users need to set up API keys like OpenAI API key, Browserbase API key and project ID. It has a simple agent loop calling Stagehand and Browserbase. Key technologies include Browserbase, Stagehand, Next.js and OpenAI. Contributions are welcome and it's licensed under MIT.
RAT-retrieval-augmented-thinking
RAT-retrieval-augmented-thinking
TLDR: RAT is a tool that enhances AI responses by leveraging DeepSeek's reasoning capabilities and combining it with various response models like OpenRouter. It offers features like model selection, reasoning visibility, and context awareness.
yapsearch
yapsearch
TLDR: The project yapsearch aims to add search and reasoning capabilities to the agent within yapthread (app.yapthread.com).
DeepSeek-R1
DeepSeek-R1
TLDR: This repository introduces DeepSeek-R1, a first-generation reasoning model. It includes details about the development of DeepSeek-R1-Zero and DeepSeek-R1 through reinforcement learning and related pipelines. The model downloads, evaluation results on various benchmarks, chat website, API platform, local running instructions, license, and citation are also provided. Multiple models are open-sourced to support the research community.
WebRover
WebRover
TLDR: WebRover is an autonomous AI agent that uses advanced language models and web automation tools to navigate the web, gather information, and provide structured responses. It has features like AI-powered navigation, smart element detection, visual feedback, and autonomous operation. It has a backend based on Python, LangChain, Playwright, OpenAI GPT-4, and FastAPI, and a frontend based on Next.js, TypeScript, Tailwind CSS, and Framer Motion. It requires setting up environment variables and has instructions for both backend and frontend setup. Contributions are welcome and it is licensed under the MIT License.
Riona-AI-Agent
Riona-AI-Agent
TLDR: Riona-AI-Agent is an AI-powered automation tool for Instagram. It can generate engaging content, automate interactions like posting, liking, and commenting. Also supports proxy and cookie management. Future features include Twitter and GitHub automation.
evabyte
evabyte
TLDR: This repository focuses on EvaByte, a 6.5B byte-level language model. It has an improved architecture with multibyte prediction and an efficient attention mechanism EVA. Trained on 1.5T bytes of data, it performs well in coding tasks and decodes fast. The repo provides model implementation based on Huggingface `transformers` library and inference examples, including different generation modes and relevant notes and limitations. Evaluation methods and citation information are also provided.
opensource_notebooklm
opensource_notebooklm
TLDR: An open-source implementation of NotebookLM using Deepseek-V3 and PlayHT TTS. Features include AI-powered conversations, text-to-speech conversion, interactive podcast-style conversations, and educational content generation. Requires Python 3.x, FAL API key, and OpenRouter API key.