#python #glm #image2text #ocr

GLM-OCR is a top 0.9B-parameter model for accurate OCR on complex documents like tables, code, formulas, seals, and receipts, scoring 94.62 on OmniDocBench V1.5. Install via `pip install glmocr`, use cloud API (no GPU needed) or self-host with vLLM/SGLang for fast, low-cost inference, and get JSON/Markdown outputs easily via CLI or Python. You benefit from quick, robust document parsing that saves time, cuts compute costs, and integrates simply into your apps for real-world tasks.

https://github.com/zai-org/GLM-OCR

#python #apple_silicon #florence2 #idefics #llava #llm #local_ai #mlx #molmo #paligemma #pixtral #vision_framework #vision_language_model #vision_transformer

MLX-VLM lets you run, chat with, and fine-tune Vision Language Models (VLMs) plus audio/video models on your Mac using MLX—install easily with `pip install -U mlx-vlm`. Use CLI for quick text/image/audio generation (e.g., `mlx_vlm.generate --model ... --image photo.jpg`), Gradio UI for chats, Python scripts, or a FastAPI server with OpenAI-compatible endpoints supporting multi-images/videos. Features like TurboQuant cut KV cache memory by 76%, and LoRA/QLoRA fine-tuning works on consumer hardware. You benefit by experimenting with powerful multimodal AI locally—fast, memory-efficient, no cloud costs, perfect for Mac users tweaking models affordably.

https://github.com/Blaizzy/mlx-vlm

#python

PersonaPlex is a real-time speech model for natural, low-latency conversations. Control its voice with audio prompts and role via simple text—like a friendly teacher, customer service rep, or casual chat partner—with natural male/female voices. Install easily, launch a web demo server, and test offline. You benefit by creating personalized AI interactions for apps, role-play, or fun talks, with quick setup and low GPU needs via CPU offload.

https://github.com/NVIDIA/personaplex

#python #ai_agents #ai_tutor #clawdbot #cli_tool #deepresearch #interactive_learning #large_language_models #multi_agent_systems #rag

DeepTutor v1.0.0 is an open-source AI tutoring tool with personalized TutorBots, unified chat modes for solving problems, quizzes, research, and math animations, plus knowledge bases from your PDFs, persistent memory of your learning style, AI co-writing, and guided plans—all via easy web, Docker, or CLI setup. You benefit by getting a smart, evolving study companion that adapts to you, boosts understanding with interactive tools, and saves time on tough topics without starting over.

https://github.com/HKUDS/DeepTutor

#python #android #android_apps #f_droid #foss #free_and_op #free_and_open_source #izzyondroid #open_source #opensource

This list offers free open-source Android apps for every need—from browsers, cameras, and music players to games, tools, and F-Droid stores like Neo-Store. Get them via F-Droid or IzzyOnDroid for no ads or tracking. It boosts your privacy and freedom by ditching Google apps, letting you control your data and phone fully.

https://github.com/offa/android-foss

#python #ai #deep_learning #filetype #keras_classification_models #keras_models #mime_types #onnx

Magika is a fast AI tool from Google that detects file types with ~99% accuracy across 200+ formats, using a tiny model that works in milliseconds on one CPU. Install easily via pip, brew, or scripts for CLI/Python/JS/Go use; scan files, directories, or streams with options like JSON output or recursion. It boosts your safety by routing files to scanners, like in Gmail/Drive, helping spot threats quickly without size limits.

https://github.com/google/magika

#python #ai_agent #automation #autonomous_agent #browser_automation #claude #computer_control #desktop_automation #gemini #lightweight #llm_agent #memory_system #python #self_evolving #skill_tree #task_automation

GenericAgent is a simple 3K-line AI agent framework that controls your computer—browser, files, mouse, screen, and phone—with just 9 tools and a 100-line loop. It learns from tasks like ordering food, checking stocks, or sending messages, saving them as reusable skills that grow into your unique skill tree over time. Install easily with git clone, pip, and an API key, then launch. This saves you hours on repetitive work, automates personal tasks, and builds smarter help tailored just for you.

https://github.com/lsdefine/GenericAgent

#python

DFlash is a lightweight block diffusion model that speeds up large language models like Qwen3.5 and Llama through speculative decoding, generating draft tokens in parallel for over 6x faster inference with no quality loss—up to 2.5x better than top methods. It supports easy integration with vLLM, SGLang, Transformers, or MLX via simple installs and commands, with ready models on Hugging Face. You benefit by running quicker AI generation on your hardware, boosting throughput to ~430 tokens/second and GPU use over 90% for efficient tasks like math or coding.

https://github.com/z-lab/dflash

#python #ai_sre #alerting #datadog #grafana #incident_management #observability #remediation #root_cause_analysis #site_reliability_engineering #slack #sre

OpenSRE is a free open-source tool to build AI agents that fix production issues fast. It connects to 40+ tools like Kubernetes, Datadog, Slack, and LLMs, then auto-fetches alerts, analyzes logs/metrics/traces, finds root causes with evidence, suggests fixes, and posts updates. Install easily with one command, run tests, and customize workflows on your infrastructure. This saves you hours on incident debugging, cuts downtime, and predicts failures—letting you focus on building instead of firefighting.

https://github.com/Tracer-Cloud/opensre

#python #bloomberg_terminal #contributions_welcome #finance #financial_markets #foss #good_first_issue #help_wanted #investing #investment #investment_research #machine_learning #opensource #python #quantitative_finance #stock_market #stocks

Fincept Terminal v4.0.2 is a free, open-source (AGPL-3.0) native C++20 desktop app with Qt6 UI and Python analytics, offering CFA-level tools like DCF models, portfolio optimization, risk metrics, 37 AI agents (Buffett-style to geopolitics), 100+ data connectors (Yahoo, FRED, brokers), real-time crypto/equity trading via 16 brokers, QuantLib suite, node editor workflows, and global intelligence. Download installers for Windows, Linux, macOS from GitHub releases or build easily with scripts/Docker. It benefits you by delivering Bloomberg-class performance in one fast binary, unlimited data access, and pro analytics to boost trading decisions and research without limits or high costs.

https://github.com/Fincept-Corporation/FinceptTerminal

#python #aigc #comfyui #image_generation #tts #video_generation

Pixelle-Video is an AI tool that fully automates short video creation: input a theme, and it writes scripts, generates images/videos, adds voiceovers, music, and compiles the final clip—no editing skills needed. Recent updates include action transfer, multi-language TTS, and custom uploads for personalized results. Download the Windows package or install via source for easy web use with free local options. You benefit by quickly making pro videos for social media or content, saving hours of manual work and costs.

https://github.com/AIDC-AI/Pixelle-Video

#python

ML Intern is an autonomous AI agent that researches, writes, and deploys high-quality machine learning code using Hugging Face tools, docs, datasets, and cloud resources. Install easily by cloning the GitHub repo, running `uv sync`, and setting API keys in a `.env` file; use interactively with `ml-intern` or headless like `ml-intern "fine-tune llama on my dataset"`. This saves you hours of manual ML work, letting you quickly prototype and ship projects without deep expertise.

https://github.com/huggingface/ml-intern

#python

Free Claude Code lets you use Claude Code CLI and VSCode for free without an Anthropic API key. It proxies requests to free providers like NVIDIA NIM (40 req/min), OpenRouter free models, DeepSeek, or local LM Studio/llama.cpp. Just set 2 env vars, edit .env with your keys/models, run the server on port 8082, and point Claude to it—works seamlessly with thinking tokens, tool parsing, and rate limits. You save money on API costs, get unlimited local options for privacy/offline coding, and mix providers for top performance.

https://github.com/Alishahryar1/free-claude-code

#python #api #ddgs #dht #mcp #mcp_server #metasearch #p2p #python #search #websearch

DDGS is a Python library (version 3.10+) for metasearch that pulls text, images, videos, news, and books from engines like Bing, DuckDuckGo, and Google. Install with `pip install ddgs`, use `DDGS().text("query")` for fast results, or run API/CLI servers for web access; optional DHT (Linux/macOS) shares anonymous caches to cut latency 90% and dodge rate limits. It benefits you by delivering quick, diverse, aggregated search data without single-engine restrictions or slow repeats—ideal for apps, scripts, or tools needing reliable info fast.

https://github.com/deedy5/ddgs

#python

Universal Commerce Protocol (UCP) is an open standard that lets AI agents, apps, businesses, payment providers, and others communicate seamlessly for secure checkouts, order updates, and personalized shopping without custom integrations. Key features include dynamic capability discovery, composable modules like Checkout and Identity Linking, transport flexibility, and strong security. You benefit by easily building agentic commerce, automating discoveries and transactions to boost efficiency, cut integration costs, and deliver better customer experiences across platforms.

https://github.com/Universal-Commerce-Protocol/ucp

#python #awesome #awesome_lists #awesome_resources #codex #codex_cli #codex_skills #coding_agent_skills #coding_agents #gpt_5_1_codex #gpt_5_codex #llm #skills

Awesome Codex Skills offers a curated list of practical skills to automate workflows in Codex CLI and API, like code reviews, meeting summaries, GitHub PR fixes, Notion organization, email drafting, and data analysis across 1000+ apps via Composio. Install easily with a Python script or manually by copying folders to ~/.codex/skills, then restart Codex—it auto-triggers skills matching your task description. This saves you hours on repetitive dev, productivity, and comms tasks, letting Codex take real actions like posting to Slack or creating issues for faster, smarter work.

https://github.com/ComposioHQ/awesome-codex-skills

#python #adb #android #android_mods #no_root #rootless #shizuku

Shizuku lets normal apps use powerful system features on non-rooted Android devices via ADB, with this curated list showcasing apps for automation (like Tasker), file management (MiXplorer), customization (DarQ), privacy, audio tweaks, and more across categories like AI agents and gaming. It helps you unlock advanced controls, boost productivity, enhance privacy, and customize your phone without rooting, saving time and extending device capabilities safely.

https://github.com/timschneeb/awesome-shizuku

#python #desktop_pet

DyberPet is a desktop pet development framework built on PySide6 that lets you create and customize animated companions for your computer screen. The framework provides developers with core tools to build desktop pets with features like customizable animations, interactive behaviors, productivity tools (Pomodoro timers, to-do lists), and AI chat capabilities using your own OpenAI API key. You benefit from a flexible, open-source platform that supports multiple pets, themes, and personalization options—whether you want a simple animated companion to brighten your workspace or a fully-featured pet that helps you stay focused and productive while working or studying.

https://github.com/ChaozhongLiu/DyberPet

#python #agent #ai #aigc #gemini #llm #quant #quantitative_trading #rag #stock

This AI-powered stock analysis system uses large models to automatically analyze your chosen A-shares, HK, or US stocks daily, delivering a decision dashboard with core conclusions, scores, buy/sell points, risk alerts, and checklists via WeChat, Telegram, email, or more. Deploy it free in 5 minutes via GitHub Actions—no server needed—or run locally/Docker. You save hours of manual research, get multi-dimensional insights (technicals, news, funds), backtesting, and interactive Agent queries for smarter, automated trading decisions.

https://github.com/ZhuLinsen/daily_stock_analysis

#python #academia #anthropic #arxiv #brave #deep_research #encryption #home_automation #homeserver #local #local_deep_research #local_llm #mistral #ollama #openai #pubmed #research #research_tool #retrieval_augmented_generation #searxng #self_hosted

Local Deep Research is a free, open-source AI tool you run locally for private, deep research on any topic. It auto-searches the web, academic papers (arXiv, PubMed), and your documents using LLMs like Ollama or GPT, then synthesizes cited reports in minutes. Install easily via Docker or pip, build an encrypted knowledge base from downloads, and get 95% accuracy. Benefits: total privacy (no tracking), zero cost for local models, customizable strategies, and compounding knowledge—saving hours on complex queries while owning your data.

https://github.com/LearningCircuit/local-deep-research