🤙 First time presenting at GITEX about sustainability & AI with OSS perspective! IMHO sustainable AI = inclusive AI, that's why Ant Open Source ❤️ InclusionAI, a suite of AI OSS projects, built by everyone, for everyone. Glad to carry the tshirt from #ICLR25 to #gitexasia! 🫠🫶

deepseek-ai/DeepSeek-Prover-V2-671B · Hugging Face huggingface.co/deepseek-ai/De…

🤩 Second time joining the #2050 event at Hangzhou and it is extremely fun. I really love what Datawhale has been doing with the “learning festival” and brought 2 sessions for sharing. Met friends old and new. Both me and my Capricorn boy love the learning vibe here! 🤗🤩

🤗Check out my latest article: Let's focus on the real recipe of authentic cuisines while enjoying packaged food - A reflection on GITEX Asia Sustainable AI panel discussion linkedin.com/pulse/lets-foc… via LinkedIn 😺 It was long reflection on my latest GITEX ASIA experience~

First time participating as a member of Program Committee for such a professional event with The Linux Foundation, glad to support the community and help everyone with high quality content 😀 credly.com/badges/de13732…

RL training is too slow? AReaL by Ant Research introduces an asynchronous approach that decouples generation from training to eliminate blocking. Combined with system-level optimizations, this method achieves up to 2.57× speedup. Code open-sourced Yi Wu

We release fully async RL system AReaL-boba² for LLM & SOTA code RL w. Qwen3-14B! Qwen #opensource 🚀system&algorithm co-design → 2.77x faster ✅ 69.1 on LiveCodeBench 🔥 multi-turn RL ready 🔗 Project: github.com/inclusionAI/AR… 📄 Paper: arxiv.org/pdf/2505.24298 1/3👇

🚀This is the year for RL system and #AReaL keeps pushing on the frontier, this time with async capability. What do you want next? Better development exp? Better usability? Just name a few~🚅

Please check out our Ming-Omni Technical Report: arxiv.org/abs/2506.09344

最近鼓吹这个药丸那个要被 AI 取代的声音甚嚣尘上，首当其冲的就是 AI Coding 替代程序员一说。 总体看下来是对选择大于努力的巨大恐惧，唯恐自己选错对象万劫不复 .. 以我浅薄的经验看，选择大于努力不假，但是只要大方向不出大问题，努力是有得选的前提。否则机遇砸到面前，接不住岂不更捶胸顿足。

Hugging Face is the Github for software 2.0 by Andrej Karpathy

Will speak at #OSSummit North America, June 23-25 in Denver. It would be a fun and engaging events with many "interesting" #OpenSource news. Explore the schedule: bit.ly/4lGUwbB I'll be at LF AI & Data Foundation booth on day 1 with a panel in the aftn. Available otherwise! 😺

Knows the team lead for years, a quiet and deep thinker who is really dedicated to #KnowledgeGraph and #MultiModal works. #MingOmni is a good model AFAIK

🥲 finally got to Denver #OSSummit NA after 25 hours of air travel and unlimited wait time. 👏 That said, looking forward to the keynotes and technical conversations tomorrow!

+1 for "context engineering" over "prompt engineering". People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window

😁 It is such a delight to see the best of the best leading the most important open cloud infrastructure work on this planet. Knowing Jonathan and Chris for years, I feel confident and we are already discussing next round of strategies 😺! CNCF OpenInfra Foundation

Had an incredibly valuable dinner tonight with Richard Bian . He showed up with one hand holding his laptop, the other still on a call. Despite a packed schedule, he gave me a full hour of thoughtful, honest sharing. Deeply appreciated. 🙏

#Supabase is the biggest winner behind this round of Coding Agent.

Tired intricate system code for RL training? 🤯 We release AReaL-lite – A lightweight AReaL version for AI researchers! 🚀#opensource ✨ Algorithm-first design & APIs🎉 ✨ 80% less code w. 90% AReaL's full efficiency 🎉 ✨ Customizable agentic RL🎉 🔗 github.com/inclusionAI/AR…

🔆A real light to the RL system options! Make RL easy to write has always been a pain point👏 This will indeed help all the researchers and phd students in this domain! GitHub github.com/inclusionAI/AR… #ReinforcementLearning #rlresearcher #phdstudent