GitHub has introduced the second cohort of GitHub Accelerator, which is a program that funds and helps chosen open-source initiatives.
Every venture receives $400,000 in help, damaged down into $40,000 in non-dilutive sponsorship funding from GitHub Sponsors and $350,000 in Microsoft advantages, reminiscent of Azure and OpenAI credit, free Copilot and GitHub initiatives, and connection to GitHub Fund and Microsoft’s M12 Enterprise Fund.
“GitHub is on a mission to allow a world the place 1 billion people name themselves builders. GitHub Accelerator is certainly one of many doable methods to help open supply—however that influence is actually made when everybody helps the expertise they depend on. We need to be sure that open supply thrives, builders have the selection to contribute full-time to the initiatives they care about most, and people relying on open supply profit from the innovation and sustainability. Collectively, we will make investments, construct, and nurture extra open supply classes, initiatives, and other people,” GitHub wrote in a weblog put up.
This most up-to-date cohort was themed round open-source AI, and the successful initiatives embrace:
- UnslothAI, which makes use of rising applied sciences to wonderful tune open supply fashions quicker and utilizing much less reminiscence than opponents
- Giskard, a library for testing and evaluating LLMs
- A-Body, which integrates AI workflows into the creator of AR/VR experiences
- Nav2, an autonomous cellular robots navigation answer
- OpenWebUI, a UI for operating AI and LLMs regionally
- LLMware.ai, which is a set of instruments that can be utilized to construct enterprise LLM functions
- LangDrive, a framework for coaching and deploying LLMs through an API and configuration information
- HackingBuddyGPT, which supplies AI copilots for safety groups
- Internet-Test, which supplies safety insights into web sites, infrastructure, and servers
- Marimo, a pocket book for AI and machine studying
- Talkd.ai, a unified LLM Chat API
