Open Source · GitHub · AI
LIVE · Jun 01, 07:13 PMGitHub RadarTrending AI repositories
Track the most relevant open source artificial intelligence repositories on GitHub. Ranked by stars, new trending projects, organized by category and language.
By Luis Fernando Roquette • Week of May 31, 2026
GitHub Radar is SWEN.live's real-time monitor of trending open source artificial intelligence repositories on GitHub. It combines the official github.com/trending algorithm with the GitHub Search API to show which LLM, agent, MCP, RAG, computer vision and audio generation projects gained the most stars in the past week, month or year.
Ranking
Top Repositórios
| # | Repositório | ★ Estrelas / Δ período |
|---|---|---|
| 1 | ||
| 2 | ||
| 3 | ||
| 4 | ||
| 5 | ||
| 6 | ||
| 7 | ||
| 8 | ||
| 9 | ||
| 10 |
Highlight of the Week
#1 repository on github.com/trending this week
harry0703
MoneyPrinterTurbo
利用AI大模型，一键生成高清短视频 Generate short videos with one click using AI LLM.
Python76.7k★
Trending
New and trending projects in the last 28 days
Top 10 by Stars
AI repositories with the most stars on GitHub (all time).
By Category
Explore AI projects by application type
By Language
Discover the best AI projects by your favorite language
Methodology
How GitHub Radar works
The SWEN.live GitHub Radar monitors trending artificial intelligence repositories on GitHub in real time, combining two sources: direct scraping of github.com/trending— GitHub's own official algorithm — and the GitHub Search API for topic filtering. Weekly, the fastest-growing AI projects include language models (such as LLaMA, Qwen and Mistral variants), autonomous agent frameworks, code generation tools and computer vision libraries. The most relevant indicator is the star delta for the period: how many stars the repository gained specifically in that week or month — data that the standard GitHub Search API does not provide. Python dominates the AI ecosystem, followed by TypeScript (apps and interfaces), Rust (high-performance inference) and Go (AI infrastructure). Data is updated every 30 minutes and available at swen.live/github-radar.
| Theme | Source | When to use |
|---|---|---|
| All, languages (Python, Rust, Go, TS, Java, C++, Julia) | github.com/trending | When GitHub already offers curation by language or general |
| LLM, Agents, MCP, RAG, Vision, Audio, Diffusion | GitHub Search API + topic:* | When you need to filter by topic — not supported by official trending |
Data sources
- ▸github.com/trending — direct scraping of GitHub's official algorithm for “All” themes and by language. Returns the repositories that GitHub itself considers trending, with the real delta of stars gained in the period.
- ▸GitHub Search API — topic filtering (
topic:llm,topic:ai-agents,topic:mcp-serveretc.) for specialized AI themes. GitHub Trending does not support topic filtering, hence the hybrid approach.
What each indicator means
- ★Total stars — cumulative number of favorites on GitHub, a proxy for popularity and community trust over time.
- +Period delta — stars gained specifically in the selected period (week or month). The most relevant indicator of real and viral growth. Available only for scraper results.
- ▸Update frequency — 30-minute ISR on Vercel + hourly cron for cache invalidation. The “LIVE” indicator at the top shows the last revalidation time.
Why monitor AI repositories
The open source AI ecosystem is one of the most dynamic in software history. Market-defining tools — such as LangChain, Ollama, LlamaIndex, Hugging Face Transformers and Stable Diffusion — emerged and gained tens of thousands of stars in weeks. Tracking this movement in real time allows developers, researchers and product leaders to identify new technologies before they go mainstream, assess the health of the open source ecosystem and find projects worth integrating or contributing to. The SWEN.live GitHub Radar was created to be the go-to reference for the tech community in this continuous monitoring.
FAQ
Frequently Asked Questions
What is SWEN.live GitHub Radar?
GitHub Radar is SWEN.live's real-time monitor of trending open source artificial intelligence repositories on GitHub. It combines the official github.com/trending algorithm with the GitHub Search API to show which LLM, agent, MCP, RAG, computer vision and audio generation projects gained the most stars in the past week, month or year.
How does GitHub Radar identify trending repositories?
The system uses two complementary strategies. For "All" themes and by language (Python, TypeScript, Rust, Go, Java, C++, Julia), it scrapes github.com/trending directly — GitHub's own algorithm. For specialized themes (LLM, Agents, MCP, RAG, Vision, Audio, Diffusion), it uses the GitHub Search API with topic:* filters sorted by stars. GitHub indexes all repositories; the radar simply reads their curated results.
How often is GitHub Radar updated?
The page is regenerated via Vercel's Incremental Static Regeneration (ISR) every 30 minutes. A daily cron at 5 AM UTC takes a star snapshot of topic repositories in Supabase, feeding the growth calculation (period delta) for LLM, Agents, MCP, RAG and other specialized themes.
What AI categories does GitHub Radar cover?
The radar covers LLMs (large language models), Agents (autonomous agent frameworks), Code (code generation and copilots), Vision (computer vision and image generation), Audio (text-to-speech and voice recognition), MCP (Model Context Protocol servers), RAG (retrieval-augmented generation) and Diffusion (stable-diffusion and generative models). It also segments by language: Python, TypeScript, Rust, Go, Java, C++ and Julia.
How does the star growth calculation (+N stars) work?
For themes scraped from github.com/trending, the delta comes directly from GitHub ("N stars this week/month" field). For themes based on the GitHub Search API (specialized topics), the calculation is done by SWEN.live: a daily cron records the star count in Supabase; the delta is the difference between the current value and the oldest snapshot within the selected period. If a repository lost stars, the delta is displayed as zero — never negative.