WOW Thanks Zhihu for translating the good story of Yibo😍🥰🤓 Interested in what we do? Contact us「[email protected] 」Let's connect and explore cutting-edge tech together!🚀
🏆 SIGCOMM 2025 Test of Time Award goes to Zhu Yibo (@StepFun_ai Co-founder) for paper "Congestion Control for Large-Scale RDMA Deployments" that introduced DCQCN, now used in systems like GPT-4.5. Zhu also shared his behind-the-scenes story on Zhihu: In summer 2013, after my…
🔥🔥🔥Big news from SIGCOMM 2025 in Portugal! Our co-founder and VP, Yibo Zhu, has won the prestigious Test of Time Award for the paper "Congestion Control for Large-Scale RDMA Deployments," which introduced the DCQCN protocol. This award celebrates groundbreaking work with…
🫣Look who's number one on the list? Oh, it's me!☺️ 🔥StepAudio 2 has topped the official MMAU leaderboard🏆 Come and check it out! sakshi113.github.io/mmau_homepage/…
We are trending fast on HuggingFace🔥🔥🔥 Thank you for liking our model🥰Fuel us with more LIKES and STARS to do better🫡🫡🫡 🔗 GitHub: github.com/stepfun-ai/Ste… 🔗 Huggingface：huggingface.co/stepfun-ai/Ste…
刚写周刊的时候发现阶跃开源的这个 Step Audio 2-7B 端到端语音模型有点意思。 应该是第一个支持推理的端到端语音模型，支持本地部署应该比 Open AI 的 GPT-realtime 便宜很多，发布这几天已经在 huggingface trending 榜上了。…
大模型能看懂电视剧吗? 给大家带来阶跃星辰 Step-Audio-2-Mini 端到端语音大模型测试! 本次测试包含汉语, 英语 ASR, 英译汉(而且是印度腔英语), 以及音频理解和推理测试. 这次使用了几个经典剧集的名场面进行测试, 包括 《绝命毒师》《生活大爆炸》《雍正王朝》《Rick & Morty》. 直接说结论:…
StepFun AI Releases Step-Audio 2 Mini: An Open-Source 8B Speech-to-Speech AI Model that Surpasses GPT-4o-Audio Step-Audio 2 Mini, released by StepFun AI, is an open-source 8B speech-to-speech model that outperforms GPT-4o-Audio in benchmarks. Trained on 8M+ hours of multilingual…
Thanks to Zhihu and HuggingFace for organizing this. Let's look forward to the wonderful discussions tonight🤓👏🔥
🚨 AI Insight Talk: HF Papers Live｜VLM Session 📅 Aug 21, 8:00 PM (GMT+8) — Live 🔥China’s Open-source VLMs boom—Intern-S1, MiniCPM-V-4, GLM-4.5V, Step3, OVIS 👨🔬Devs from Shanghai AI Lab, @OpenBMB , @Zai_org , @AlibabaGroup and @StepFun_ai will share their…
Come to HuggingFace space and try it out🙌
🚀 Introducing NextStep-1: A new paradigm for autoregressive image generation. 💡 Core design: A 14B Causal Transformer "artist" paired with a lightweight 157M Flow Matching "brush". We generate directly in continuous visual tokens, bypassing the information bottleneck of…
🚀 Exciting release: StepFun-Prover-Preview-7B & 32B—LLMs for formal theorem proving via tool-integrated reasoning! Using RL + environment feedback, they generate Lean 4 proofs efficiently. 🔥 7B matches DeepSeek-Prover-V2-671B & Kimina-Prover-72B on miniF2F-test (pass@1)! 💡 32B…
2️⃣ In Agent scenarios with ultra-long context (128K+), A/E separation becomes way more important. 3️⃣ Why bother with A/E separation? ➤ Hardware optimization → Match A&E to different hardware for max throughput per cost. ➤ Utilization boost → Even without buying new hardware,…
As promised, StepFun has released the Step 3 model on Hugging Face: huggingface.co/stepfun-ai/ste…
The wait is over! Meet Step 3 — the groundbreaking multimodal LLM from StepFun! 🚀 MoE architecture (321B total params, 38B active) 💡 Rivals OpenAI o3, Gemini 2.5 Pro, and Claude Opus 4 in performance 🖥️ Optimized for China’s domestic AI chips StepFun just announced: Step 3…
United States 趨勢
- 1. D’Angelo 21.7K posts
- 2. Angie Stone N/A
- 3. Happy Birthday Charlie 89.5K posts
- 4. #BornOfStarlightHeeseung 59.5K posts
- 5. Brown Sugar 2,155 posts
- 6. #csm217 1,811 posts
- 7. #tuesdayvibe 5,190 posts
- 8. Alex Jones 20.1K posts
- 9. Black Messiah N/A
- 10. Sandy Hook 6,522 posts
- 11. Pentagon 85.6K posts
- 12. Drew Struzan N/A
- 13. #NationalDessertDay N/A
- 14. #PortfolioDay 5,672 posts
- 15. Cheryl Hines 1,718 posts
- 16. George Floyd 6,201 posts
- 17. Taco Tuesday 12.6K posts
- 18. Masuda 2,482 posts
- 19. Monad 218K posts
- 20. Powell 20.3K posts
Something went wrong.
Something went wrong.