Hi Community: Over the last year, we have witnessed a great number of innovations with the arrival of foundational models, including stable diffusion models for image generation, whisper for voice recognition, GPT, and open LLMs(llama2, MPT, Falcon, RedPajama).
As of now, the unity is being developed as a branch for GenAI support. Because of the level of new features being brought in, as well as the ML/AI landscape shift we are experiencing. As of now, the unity branch includes all the main features, with additional modules to support genAI-related needs. The community held quite a few conversations about our general GenAI strategy. Most of the community agrees that it is important to bring unity support officially in a timely manner. As of now we also start to see the majority of new development activities shift to the unity branch, making it the de-facto main. TVM unity is a major change in terms of the set of approaches to reflect the trends in AI/ML ecosystem. We have also developed the system in a way that involves continuous communication with the community for over a year with various applications that are already being used in LLM use cases. After discussions with the community members, this post proposes to take the path that most favor: change the current codebase main branch to the unity branch, ensuring the main modules are incorporated (ensure git merge of latest main into unity). This is a formal voting thread; The vote will be open for at least one week, please vote: +1 if you support the transition +0 -1 because -- Reply to this email directly or view it on GitHub: https://github.com/apache/tvm/issues/16368 You are receiving this because you are subscribed to this thread. Message ID: <apache/tvm/issues/16...@github.com>
