Log inSign up
Song Han
380 posts
Song Han profile banner
@songhan_mit

Song Han

@songhan_mit
Joined March 2019
240
Following
10.5K
Followers
RepliesRepliesRepostsRepostsMediaMedia

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
  • @songhan_mit
    Song Han
    @songhan_mit
    Sep 4
    Ten years, same question, much harder version of it. Jun-Yan has been pushing the frontier of visual generation — someone I’ve admired since our college days. Now it’s @NunchuxAI. Congratulations!
    @junyanz89
    Jun-Yan Zhu
    @junyanz89
    Sep 3
    Nunchux AI (nunchux.ai) comes out of stealth today. Ten years ago at Berkeley, we spent whole nights on a Titan X in Theano, trying to make iGAN, pix2pix, and CycleGAN real-time so creators could see the results the moment they asked for them. Image models now
    1
  • @songhan_mit
    Song Han
    @songhan_mit
    Sep 3
    Seven years ago, Muyang started working on this problem before it was popular. Zhekai and Yujun have been building alongside him ever since, and together with Jun-Yan — one of the sharpest people I know in this field — they’ve turned it into @NunchuxAI. Congrats to all of them.
    @lmxyy1999
    Muyang Li
    @lmxyy1999
    Sep 3
    Today is a meaningful day for me. Seven years ago, as an undergraduate, I started researching efficient generative AI. At the time, few people saw generative AI becoming something people would use every day. GANs were still the frontier. Then came diffusion. Now we are beginning
    1
  • @songhan_mit
    Song Han
    @songhan_mit
    Sep 3
    Glad to see HAN Lab Mafia keep pushing the frontier
    @zhijianliu_
    Zhijian Liu
    Inco AI
    @zhijianliu_
    Sep 3
    Hmm, another inference company topping the leaderboard? Oh wait, that's us! Speed of light on Artificial Analysis for four major open-source models. ⚡️
    3
  • @songhan_mit
    Song Han
    @songhan_mit
    Sep 3
    Amazing innovation and execution. Congrats!
    @NunchuxAI
    Nunchux AI
    @NunchuxAI
    Sep 3
    Nunchux is live at nunchux.ai! We’re building the frontier of multimodal generative AI inference: fast, affordable, high-quality serving for image, video, and world models. Our Modelverse brings 30+ image, video, and avatar models behind one API. Use
    Image
  • @songhan_mit
    Song Han
    @songhan_mit
    Sep 3
    This is astonishingly fast. Congrats!
    @inco_ai
    Inco AI
    @inco_ai
    Sep 3
    Inco AI is entering public beta with the ⚡️ fastest ⚡️ #inference endpoints for #Kimi K3, #MiniMax M3, #GLM 5.3, and #GLM 5.3 Flash. Ranking Number 1 on the respective @ArtificialAnlys provider leaderboard for output speed. Read more: inco.ai/blog/inco-plat…
Advertisement
Advertisement