Othersnotebooklm-pyPWnotebooklm, google, python, api, agent
notebooklm-py
Google NotebookLM's unofficial Python SDK supports CLI and AI Agent integration, providing programmable access to all NotebookLM features.
6074
View repositoryIndependent Column
Open-source projects worth exploring
Runnable, documented projects worth studying and extending.
Othersnotebooklm-pyPW
OthersHermes AgentPWagent, nous-research, python, llm
Hermes Agent
NousResearch's intelligent agent framework continuously evolves with use, supporting tool calls and multi-turn conversations.
6072
View repositoryOthersQwen-AgentPWAgent framework, Qwen, MCP, Function Calling
Qwen-Agent
Based on Qwen 3.0+, the Agent development framework supports Function Calling, MCP, Code Interpreter, RAG, Chrome extensions, etc., and has served as the backend for Tongyi Qianwen Chat.
6072
View repository
OthersAliGaode Maps has also developed a world model: FantasyWorld, which currently ranks first in several key indicators on the WorldScore world model leaderboardPWWorld models, models
AliGaode Maps has also developed a world model: FantasyWorld, which currently ranks first in several key indicators on the WorldScore world model leaderboard
Aligaode Maps' world model FantasyWorld ranked first on the WorldScore leaderboard, aiming to address the shortcomings of existing video models in 3D understanding through unified video generation and 3D prediction, improving spatial consistency and 3D reasoning capabilities. This model supports AR/VR content generation and robot navigation, suitable for various 3D task scenarios such as navigation, quick renovation, and e-commerce displays
6075
View repository
Efficiency toolsNVIDIA's latest open-source ASR: Nemotron Speech ASR, specialized in low-latency real-time voice agent scenarios, enabling multiple people to chat simultaneously without excessive latencyPWASR, open source
NVIDIA's latest open-source ASR: Nemotron Speech ASR, specialized in low-latency real-time voice agent scenarios, enabling multiple people to chat simultaneously without excessive latency
Nvidia's Nemotron Speech ASR focuses on low-latency real-time voice processing, supporting multi-person conversations. Each sentence can be transcribed in just 24 milliseconds, with overall latency below 500 milliseconds. Through the FastConformer architecture and cache-aware mechanism, incremental computation of voice features is achieved, dynamically configuring latency patterns to meet different application needs
6077
View repository
AI and Agentsnotebooklm-pyPWOpen source projects
notebooklm-py
Google NotebookLM is an unofficial Python API that can procedurally create notebooks, upload source files, and generate audio podcast summaries.
6079
View repository
AI and AgentsWorldMonitorPWOpen source projects
WorldMonitor
Real-time global intelligence dashboard—AI-driven news aggregation, geopolitical monitoring, and infrastructure tracking, all with a unified situational awareness interface. This week, GitHub's top trending group is 33k+ stars.
6076
View repositoryEfficiency toolsTencent has open-sourced its 3D world generation model, HunyuanWorld 1.0, which can create immersive, explorable, and interactive 3D worldsPW3D world generation, open source, modeling
Tencent has open-sourced its 3D world generation model, HunyuanWorld 1.0, which can create immersive, explorable, and interactive 3D worlds
In terms of results, the generated images are of high quality, with rich detail rendering and some physical simulation capabilities. It also supports text-to-image and image-to-image generation, offering diverse styles and generating scenes for 360° immersive roaming. Supports mesh export, seamlessly compatible with existing computer graphics pipelines
6072
View repositoryEfficiency toolsA system that reconstructs 3D human skeletal and grid models from a single image: HSMR, which can convert an ordinary photo into a 3D human model with biomechanically accurate bone structuresPW3D human model generation, model
A system that reconstructs 3D human skeletal and grid models from a single image: HSMR, which can convert an ordinary photo into a 3D human model with biomechanically accurate bone structures
HSMR systems can reconstruct biomechanically accurate 3D human skeletal and mesh models from single images, suitable for complex poses and poor shooting angles, and provide methods for creating pseudo-reality datasets to train other models.
6075
View repositoryEfficiency toolsA 3D human model generation project: StructLDM, which can generate complete 3D human bodies and supports partial editing and modifications, such as changing clothes or changing posesPW3D human body generation, model
A 3D human model generation project: StructLDM, which can generate complete 3D human bodies and supports partial editing and modifications, such as changing clothes or changing poses
StructLDM is an open-source project capable of generating complete 3D human models, supporting the generation of different poses and perspectives, as well as editing and modifying clothing and body shapes, allowing for local try-on and identity swapping.
6077
View repositoryEfficiency toolsBrowser UsePWBrowser automation, agents, web operations, AI
Browser Use
Let AI Agents browse web pages like humans, performing clicks, typing, and navigation operations
6072
View repositoryEfficiency toolsLobeHub TTSPWTTS, speech synthesis, multi-engine, open source
LobeHub TTS
An open-source, high-quality TTS toolkit supporting multiple TTS engines and voices
6078
View repositoryOthersE2BPWSandbox, secure execution, cloud infrastructure, agents
E2B
Provides a cloud infrastructure for AI Agents in a secure sandbox environment, securely executing AI-generated code
6076
View repositoryAI and AgentsDevikaPWAI engineers, Devin replacements, code generation, autonomy
Devika
An open-source AI software engineer, similar to Devin, can understand requirements and write code independently
6072
View repository
OthersDB-GPTPWData applications, databases, agents, multi-model
DB-GPT
AI-native data application development framework supporting AWEL and Agent multi-model data-driven applications
6075
View repository
AI and AgentsAutoGenPWMicrosoft, multi-agent, conversation, collaboration
AutoGen
Microsoft's open-source multi-agent dialogue framework supports automatic dialogue collaboration among agents to solve problems
6073
View repositoryOthersCogVideoPWZhipu, text-to-video production, multi-resolution, open source
CogVideo
Zhipu's open-source text-to-video model CogVideoX supports multiple resolutions and durations
6077
View repositoryEfficiency toolsCrawl4AIPWWeb crawlers, asynchronous, structured data, AI
Crawl4AI
Asynchronous web crawlers optimized for AI and LLMs, automatically extracting structured data
6075
View repositoryOthersSemantic KernelPWMicrosoft, SDK, orchestration, plugins
Semantic Kernel
Microsoft's open-source AI orchestration SDK integrates LLMs with traditional programming languages, supporting plugins and planning
6078
View repositoryEfficiency toolsStirling PDFPWPDF tools, self-hosted, editing and conversion, open source
Stirling PDF
A powerful self-hosted PDF tool supporting AI-assisted PDF editing and conversion
6071
View repositoryOthersHaystackPWNLP, RAG, search, Q&A systems
Haystack
An end-to-end NLP/LLM application framework supporting RAG, question answering, semantic search, and more
6076
View repositoryEfficiency toolsContinuePWIDE plugin, code assistant, VSCode, open source
Continue
An open-source AI code assistant IDE plugin supporting VS Code and JetBrains, capable of connecting to any LLM
6076
View repositoryOthersnotebooklm-pyPWAI, NotebookLM, API, Python
notebooklm-py
Google NotebookLM is an unofficial Python API that supports programmatic access to all NotebookLM features (including Web UI capabilities not exposed), and can be integrated with AI agents such as Claude Code and Codex.
5974
View repositoryOthersMiroFishPWAI, swarm intelligence, prediction, Python
MiroFish
A simple and universal swarm intelligence engine that can predict everything. Based on swarm intelligence algorithms, supports multi-domain predictive analytics, with a star rating of 16k+.
5970
View repository