Wan2.2 is an open-weights family of large-scale video generative models released by the Wan-Video team as the successor to Wan2.1. It introduces expanded training data and architecture upgrades, offering multiple variants for text-to-video, image-to-video, video-to-video, and even speech-to-video generation. The models produce cinematic, higher-resolution clips with stronger temporal consistency and motion control, and are available under an Apache-2.0 license with runnable repos and Hugging Face checkpoints.

Wan2.2

wan2-2

Open-source video generation model family (5B/14B) for text/image/video-to-video, delivering 1080p clips with improved motion and control.

Cell2Sentence (C2S) is an open-source framework from the van Dijk Lab that represents single-cell RNA-seq profiles as textual “cell sentences” (genes ordered by expression), enabling large language models to natively model biology. The project includes code, docs, and open models (C2S-Scale) released with Google Research/DeepMind for state-of-the-art single-cell tasks like cell typing, perturbation prediction, and virtual cell generation.

Cell2Sentence

cell2sentence

Open-source framework that turns single-cell gene expression into 'cell sentences' so LLMs can analyze and generate biology insights.

Rasa is an open-source conversational AI framework for building contextual assistants in text and voice. Teams use Rasa Open Source for NLU, dialogue, and channel connectors, and scale with Rasa Pro and Rasa Studio for enterprise features like governance, analytics, and a no-code UI. The platform supports on-prem or cloud deployment, integrates with messaging/voice channels and backends, and emphasizes data control, reliability, and extensibility for production CX use cases.

Rasa

rasa

Open-source conversational AI framework with Pro & Studio tools to build secure, production-grade chat and voice assistants.

Wildcard AI provides an open-source specification called *agents.json* that builds on OpenAPI to let agents consume APIs more reliably. It enables defining API contracts, flows (chains of calls), links between actions, and metadata so autonomous agents can discover and invoke workflows automatically. It includes Wildcard Bridge SDK, a registry of agents.json files, and motivations around statelessness, better tool-selection, and easier LLM-API integration.

Wildcard AI / agents.json

wildcard-ai-agents-json

Open-source spec & platform letting AI agents discover and invoke API workflows via agents.json rather than fixed prompts.

legacy-use is an open-source platform that wraps legacy or desktop software and exposes functionality via REST APIs. It simulates GUI interactions (e.g. via VNC, RDP) to allow AI or agentic systems to invoke operations on software that lacks built-in interfaces. It supports prompt-based endpoint generation, model-provider independence, built-in logging, debugging, and optional guardrails and enterprise deployments.

legacy-use

Open-source tool exposing legacy desktop apps via REST API so AI agents can control them via GUI automation.

LangChain Social Media Agent is an open-source AI agent built on LangGraph that takes a URL, generates Twitter and LinkedIn posts, and uses a human-in-the-loop approval flow. It supports content parsing, posting cadence (via cron jobs), Slack-triggered ingestion, and optional image handling. Users can audit and edit drafts before publishing, ensuring quality control and automation.

LangChain Social Media Agent

langchain-social-media-agent

Open-source agent that drafts, reviews, and schedules social posts from URL content with human-in-the-loop.

Vercel AI SDK is a free, open-source TypeScript library from the creators of Next.js, designed to help developers build AI-powered applications and agents. It offers a unified provider API—OpenAI, Anthropic, Claude, Google Gemini, etc.—structured output (via Zod), streaming responses, chat UI integrations, tool calling, agentic loops, and full TypeScript type safety across React, Svelte, Vue, Angular, or Node apps. With over 2 million weekly downloads, it simplifies LLM integration and enables rapid development of robust AI workflows.

Vercel AI SDK

vercel-ai-sdk

Free open-source TypeScript SDK for building LLM-powered apps and agents across frameworks with unified API.

GLM‑4.5 is an open‑source, Mixture‑of‑Experts (MoE) large‑language model from Z.ai, designed for intelligent agent applications. It features 355 B parameters (32 B active) with an efficient ’thinking/non‑thinking’ hybrid reasoning mode and 128 K context length. A smaller variant, GLM‑4.5‑Air, offers 106 B parameters (12 B active) for more efficient deployments. GLM‑4.5 excels at reasoning, coding, and tool‑calling tasks and ranks among the top performers on benchmarks compared with models like o3, Grok‑4, Claude Sonnet, and GPT‑4.1.

GLM‑4.5

glm-4-5

Open-source hybrid‑reasoning MoE foundation model optimized for intelligent agent tasks with 128K context and tool use.

Activepieces is an open-source, AI-first no-code automation platform that enables teams and enterprises to build autonomous agents and workflows via a visual builder. It supports hundreds of integrations, self‑hosting, and offers agentic capabilities through MCP servers, AI agents, and built‑in tables.

Activepieces

activepieces

Open-source, no-code AI-first automation platform to build workflows and autonomous agents.

Qwen3‑Coder is an agentic code model from Alibaba’s Qwen3 family. Its flagship variant (480B MoE, 35B active) excels at agentic coding, tool use, and browser tasks with long context (256K–1M tokens). It’s open‑source under Apache‑2.0 and integrates with Qwen Code CLI and Qwen‑Agent workflows.

Qwen3‑Coder

qwen3-coder

Open‑source Mixture‑of‑Experts code model by Alibaba for agentic coding and tool use.

TEN Framework is an open-source modular system for building autonomous AI agents capable of reasoning, memory management, and tool usage. Designed to reflect real-world constraints, it incorporates a Thought-Emission-Norm (TEN) architecture, encouraging agents to think before acting. The framework supports open-ended planning, persistent memory storage, and tool augmentation. Developers can use TEN to experiment with agent autonomy and deploy agents that simulate complex decision-making workflows.

TEN Framework

ten-framework

An open-source framework for building autonomous AI agents that interact with the real world through tools and memory.

Trae Agent is an open-source, LLM-powered command-line assistant developed by ByteDance for software development workflows. It supports multi-LLM providers, file editing, bash execution, tool integration, trajectory logging, and a modular architecture—ideal for debugging, patching, documentation, and research.

Trae Agent

trae-agent

Open-source CLI LLM agent for automating software engineering tasks via natural-language commands.

Explore open-source AI Agents