AI Agents & Automation

smolGPT offers a minimal PyTorch implementation for training small Large Language Models (LLMs) from scratch, designed primarily for educational purposes and simplicity. It boasts a pure PyTorch codebase with no abstraction overhead, incorporating modern architectural elements like Flash Attention (when available), RMSNorm, SwiGLU, and optional Rotary embeddings (RoPE). The tool supports efficient training features including mixed precision (bfloat16/float16), gradient accumulation, learning rate decay with warmup, weight decay, and gradient clipping. It also includes built-in TinyStories dataset processing and SentencePiece tokenizer training integration, making it a comprehensive yet accessible platform for learning LLM development.

DaltonTx redefines drug discovery by providing an AI-enabled platform that serves as an intelligence backbone for modern R&D. It offers an adaptive intelligent system that evolves with scientific advancements, integrates seamlessly into existing workflows, and empowers users with lasting capabilities. The platform learns from every scientist, model, and experiment, continuously improving and guiding better decisions. DaltonTx's technology covers the full discovery lifecycle, including data ingestion, model training, molecule generation, and experiment prioritization. It is built by scientists for scientists, combining software engineering, machine learning, and deep drug discovery expertise to tackle complex problems in both small molecules and biologics.

Search-R1 is an open-source reinforcement learning framework designed for training large language models (LLMs) to effectively reason and make tool calls, specifically to search engines, in a coordinated manner. Built upon the veRL framework, it extends the concepts of DeepSeek-R1(-Zero) by integrating interleaved search engine access and offering a comprehensive RL training pipeline. This framework serves as an alternative to OpenAI DeepResearch, fostering research and development in tool-augmented LLM reasoning. It supports various RL methods like PPO, GRPO, and reinforce, accommodates different LLMs such as Llama3 and Qwen2.5, and integrates with diverse search engines including local sparse/dense retrievers and online search engines like Google and Bing.

Scrapecraft is an AI-powered web scraping editor designed to simplify the creation and management of web scraping pipelines. It offers a visual workflow builder, allowing users to intuitively design their scraping processes. Leveraging AI assistance, similar to tools like Cursor but specialized for web scraping, Scrapecraft enables users to build, test, and deploy scrapers using natural language prompts. Key features include support for multi-URL bulk scraping, dynamic schema definition with Pydantic, and Python code generation with async capabilities. The platform also provides real-time WebSocket streaming for data and offers results visualization in table and JSON formats. Built with a robust tech stack including FastAPI, LangGraph, ScrapeGraphAI, React, and PostgreSQL, Scrapecraft also supports auto-updating deployments via Watchtower, ensuring continuous operation without manual intervention.

Triton Inference Server is an open-source inference serving software designed to streamline AI inferencing across various environments, including cloud, data centers, edge, and embedded devices. It supports a wide array of deep learning and machine learning frameworks such as TensorRT, PyTorch, ONNX, OpenVINO, and Python. Triton optimizes performance for different query types, including real-time, batched, ensembles, and audio/video streaming. Key features include concurrent model execution, dynamic batching, sequence batching for stateful models, and a Backend API for custom operations. It also provides HTTP/REST and gRPC inference protocols, C and Java APIs for in-process use cases, and metrics for GPU utilization and server latency. Triton is part of NVIDIA AI Enterprise, offering enterprise support.

SPO (Self-Supervised Prompt Optimization) is an AI tool hosted on Hugging Face Spaces designed to enhance the performance of language models by optimizing user prompts. It allows users to create or select templates, configure various settings, and initiate an optimization process to achieve better responses from AI models. This application is particularly useful for prompt engineers and researchers looking to fine-tune their interactions with large language models, ensuring more accurate and relevant outputs through a self-supervised learning approach. The tool aims to streamline the prompt engineering workflow, making it easier to experiment with and improve prompt effectiveness.

Semantra is a multipurpose command-line tool designed for semantic search across local documents, including text and PDF files. Unlike traditional keyword matching, Semantra allows users to query by meaning, providing a more intuitive and powerful search experience. It processes documents locally, launching a web search application for interactive querying. This tool is particularly useful for individuals needing to sift through large volumes of information, such as journalists analyzing leaked documents, researchers exploring academic papers, or students engaging with literature. Semantra prioritizes privacy and security by performing all analysis on the user's computer, and it offers configurable options for embedding models and search parameters.

Simd is a free, open-source C++ image processing and machine learning library designed for C and C++ programmers. It offers a wide array of high-performance algorithms, including pixel format conversion, image scaling and filtration, statistical information extraction, motion detection, object detection, classification, and neural network functionalities. The library is highly optimized, utilizing various SIMD CPU extensions such as SSE, AVX, AVX-512, and AMX for x86/x64, NEON for ARM, and HVX for Hexagon architectures. Simd provides both a C API and C++ classes for ease of access, supporting dynamic and static linking across Windows and Linux with MSVS, G++, and Clang compilers. It also includes a Python wrapper for broader accessibility.

File AI is an AI-native data preparation and automation platform designed to unify data capture, governance, and orchestration into auditable AI workflows. It transforms unstructured data into trusted intelligence across various enterprise functions. The platform features fileForge, an AI-native data intelligence engine, alongside purpose-built solutions like fileLedger for financial operations automation and fileShield for intelligent case management in regulated environments. Key capabilities include multimodal AI OCR, classification, schema extraction, SOP-driven workflow engines, and over 100 ERP and system integrations. File AI aims to build the foundation for agentic AI at scale, providing the context, validation, and control needed for AI agents to act with confidence in real enterprise workflows.

Seed1.5-VL is a powerful and efficient vision-language foundation model developed by the ByteDance Seed Team. It is engineered to advance general-purpose multimodal understanding and reasoning, demonstrating state-of-the-art performance across numerous public benchmarks. The model features a relatively modest architecture, comprising a 532M vision encoder and a 20B active parameter MoE LLM, yet it excels in complex reasoning tasks, OCR, diagram understanding, visual grounding, 3D spatial understanding, and video comprehension. Seed1.5-VL also shows strong capabilities in interactive agent tasks like GUI control and gameplay, making it versatile for various applications. The project provides a usage cookbook with diverse code samples to help developers effectively leverage its API.

Show Facebook Computer Vision Tags is a simple browser extension for Chrome and Firefox designed to make users aware of the automated image tagging performed by Facebook's Deep ConvNet. Since April 2016, Facebook has been adding alt tags to uploaded images, populated with keywords describing their content. This extension overlays these generated tags directly onto photos in your Facebook timeline, allowing you to see what objects, activities, locations, and events Facebook's AI identifies. While these tags improve accessibility for blind users, the extension's primary goal is to highlight the extensive data extraction capabilities of major internet companies from user photographs, prompting users to consider their digital privacy. It's a straightforward tool for anyone curious about the information Facebook gleans from their visual content.

Collate is a privacy-first AI reader designed for Mac users, enabling them to chat with, summarize, and extract insights from PDF documents entirely offline. This local-first approach ensures that all processing runs directly on your device, guaranteeing complete privacy as your documents never leave your computer. It supports both Apple Silicon (M1, M2, M3) and Intel Macs running macOS 13.1 or later. Users can ask questions, get instant summaries, and receive citation-backed answers with automatic highlighting. Collate also supports multi-PDF chat for comparative research, folder organization, and the ability to export summaries and conversations in various formats like PDF, rich text, or email. It's completely free to download and use, with no subscription fees or usage limits.

sidekick.nvim is a powerful Neovim AI sidekick designed to enhance the coding experience by integrating Copilot LSP's "Next Edit Suggestions" directly into the editor. It provides automatic suggestions, rich diff visualizations with Treesitter-based syntax highlighting, and hunk-by-hunk navigation for reviewing changes. Beyond suggestions, it features an integrated AI CLI terminal for interacting with popular AI command-line tools like Claude, Gemini, and Copilot CLI, all without leaving Neovim. The tool offers context-aware prompts, a library of pre-defined prompts for common tasks, and session persistence with tmux and zellij integration. It is highly extensible and customizable, allowing users to fine-tune configurations and integrate with other plugins.

StyleGAN-Human Interpolation is a web-based tool hosted on Hugging Face Spaces, designed for generating and manipulating human faces using AI. It leverages StyleGAN models to create realistic synthetic faces, offering users the ability to explore the capabilities of this advanced generative adversarial network. The primary function of the tool is to produce a series of images that smoothly transition between two distinct, randomly generated human images. Users can control this interpolation process by adjusting parameters such as seed values and truncation psi, which influence the randomness and realism of the generated faces. This makes it a valuable resource for researchers, artists, and enthusiasts interested in AI-driven image synthesis and the nuances of facial generation.

Cloudpick specializes in advanced unmanned retail solutions, leveraging AI and multi-dimensional sensor technology to create digital twins of physical spaces. Their offerings include AI Smart Stores for automated settlements and 24/7 operation, Moby Marts for mobile retail, and Computer Vision Coolbinets for accessible mini AI unmanned stores. The platform also features a Smart Store Management System for remote control and operational efficiency. Cloudpick provides tailored solutions for diverse sectors such as transportation hubs (railway, highway, airport), cultural/sports/tourism venues, enterprise/industrial parks, and hospitals/factories/research institutes, aiming to enhance shopping experiences, boost employee efficiency, and improve store productivity through data-driven insights.

SWE-agent is an advanced agentic framework designed to enable language models (LMs) like GPT-4o or Claude Sonnet 4 to autonomously identify and fix issues within real GitHub repositories. Beyond software engineering tasks, it can be employed for offensive cybersecurity challenges, such as capture the flag, and competitive coding. The tool is highly configurable, governed by a single YAML file, and offers maximal agency to the LM, making it free-flowing and generalizable. Developed by researchers from Princeton University and Stanford University, SWE-agent has achieved state-of-the-art results on the SWE-bench benchmark. Users can try SWE-agent in their browser or explore its capabilities for offensive cybersecurity through its EnIGMA mode.

CodeFlying is an innovative AI-powered platform designed for "vibe coding," allowing users to build full-stack applications simply by describing their ideas to an AI. This no-code solution streamlines the app development process, enabling the creation of web apps, mobile apps, and even WeChat mini-programs in minutes. It aims to democratize app creation, making it accessible to individuals without extensive coding knowledge. The platform focuses on rapid prototyping and deployment, transforming conversational input into functional applications, marking a new era in app development.

SWE-RL is an official codebase for "Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution," designed to scale reinforcement learning-based LLM reasoning for real-world software engineering tasks. It leverages open-source software evolution data and rule-based rewards to improve LLM performance. The codebase includes prompt templates and a flexible reward function API that supports various editing formats, including sequence similarity for search/replace changes and unified diffs. Additionally, SWE-RL features an Agentless Mini component for fast asynchronous inference, code refactoring, file-level localization, and repair, supporting OpenAI-compatible endpoints and Hugging Face models like Llama-3.3-70B-Instruct.

Deix S.r.l. specializes in developing innovative algorithms and applications by leveraging expertise in mathematical modeling, artificial intelligence, and optimization. They provide solutions that enable companies to make informed decisions and identify new business opportunities. Deix offers both ready-to-use products and tailor-made solutions designed to meet specific business needs. Their approach integrates internal knowledge and data to deliver high-quality, efficient results, as evidenced by client testimonials highlighting speed, technical expertise, and proactivity in solving complex challenges.

sqlite-vss is a SQLite extension designed to bring vector search capabilities directly into SQLite databases, leveraging the Faiss library for efficiency. It enables developers to build semantic search engines, recommendation systems, and question-and-answering tools by storing and querying vector embeddings. While not actively developed, with efforts now focused on sqlite-vec, it offers a robust solution for integrating vector search into applications using SQLite. Users can create virtual tables to store high-dimensional embeddings and perform k-nearest neighbor searches. It supports various languages through bindings like Python, Node.js, Deno, Ruby, Elixir, Go, and Rust, making it accessible to a wide range of developers.

Falcondale specializes in developing applied quantum machine learning and optimization solutions designed to deliver real-world impact. The company focuses on leveraging quantum intelligence to solve complex problems across various industries. Falcondale aims to provide a competitive edge through its advanced quantum technologies, offering solutions that go beyond traditional computational methods. Their expertise lies in translating cutting-edge quantum research into practical, deployable applications for businesses and organizations seeking innovative data analysis and optimization capabilities.

StreamingLLM is an innovative open-source framework designed to address the challenges of deploying Large Language Models (LLMs) in streaming applications that require processing infinite-length inputs. It introduces the concept of "attention sinks" to efficiently manage Key and Value (KV) states, allowing LLMs to generalize to infinite sequence lengths without fine-tuning. This approach prevents the performance degradation seen in traditional window attention methods when text length exceeds cache size. StreamingLLM enables models like Llama-2, MPT, Falcon, and Pythia to perform stable and efficient language modeling with millions of tokens, offering up to a 22.2x speedup over sliding window recomputation baselines. It is particularly optimized for scenarios such as multi-round dialogues where continuous operation without extensive memory or dependency on past data is crucial.

StreamingVLM is an innovative AI tool designed for real-time understanding of effectively infinite video streams. Developed by mit-han-lab, it addresses common challenges in long-video analysis by maintaining a compact KV cache and aligning training directly with streaming inference. This approach efficiently avoids the quadratic cost associated with traditional methods and mitigates the pitfalls of sliding-window techniques. The system is capable of running at up to 8 frames per second (FPS) on a single H100 GPU, offering stable and efficient video processing. It has demonstrated superior performance, winning 66.18% against GPT-4o mini on a new long-video benchmark and also enhances general Video Question Answering (VQA) capabilities without requiring task-specific fine-tuning. The project provides scripts for environment setup, inference, supervised fine-tuning (SFT), and various evaluations including OVOBench and VQA tasks.

terminal-bench is an open-source benchmark designed to evaluate the performance of AI agents, specifically Large Language Models (LLMs), in realistic terminal environments. It provides a comprehensive suite of tasks that challenge agents with complex, end-to-end scenarios, ranging from compiling code to training models and setting up servers. The tool consists of a dataset of tasks, each with an English instruction, a test script for verification, and a reference solution, along with an execution harness that connects the language model to a sandboxed terminal environment. This setup ensures reproducible and practical evaluation of system-level reasoning. It is currently in beta with approximately 100 tasks, with plans for significant expansion, and welcomes community contributions for new and challenging tasks.