Log inSign up
Image
user avatar
Eric Vishria
Benchmark
@ericvishria
GP @Benchmark Director @confluentinc @Cerebras @Contentful @Benchling @CommerceLayer @acuitymd @FireworksAI_HQ @quilterai @pomerium_io @greptile @sundayrobotics
Joined June 2008
856
Following
22.5K
Followers
RepliesArticlesMedia
  • Pinned
    user avatar
    Eric Vishria
    Benchmark
    @ericvishria
    Jan 14
    .@benchmark co-lead the initial round for Cerebras 10 years ago. Over the following 5 years, the team amazed, delivering the technological marvel of a wafer-scale chip, the system to heat and cool it, and more recently the software layer that allows giant fleets of Cerebras
    user avatar
    Andrew Feldman
    Cerebras
    @andrewdfeldman
    Jan 14
    @OpenAI and @cerebras have signed a multi-year agreement to deploy 750 megawatts of Cerebras wafer-scale systems to serve OpenAI customers. This has been a decade in the making. Deployment begins in early 2026, and when fully rolled out, it will be the largest high-speed AI
    Image
  • user avatar
    Eric Vishria
    Benchmark
    @ericvishria
    Jul 25
    Yes - energy is the issue, all this other stuff is a distraction. China is bringing online something like 10x the amount of new energy we are next year. That's cheaper tokens, cheaper (and more) RSI, cheaper manufacturing, ...
    user avatar
    Chamath Palihapitiya
    The All-In Podcast
    @chamath
    Jul 25
    Here is what all these sideline commentators don’t understand: THERE IS NO PRICE DUMPING IN OPEN SOURCE AI China can price dump in manufacturing. Why? Because we destroyed our means of equivalent production and so can’t do the same. But manufacturing is not AI. China CANNOT
  • user avatar
    Eric Vishria
    Benchmark
    @ericvishria
    Jul 20
    Yes, AND, more immediately, if we ban Chinese models, American companies will be 100% reliant on OAI and Anthropic to protect ourselves. Meanwhile, North Korean & Iranian hackers will use Chinese models relentlessly to hack our government and companies. See @huggingface
    user avatar
    Chamath Palihapitiya
    The All-In Podcast
    @chamath
    Jul 20
    The leading AI has already forked into two options. A: Closed source American that costs $26-56 per 1MM tokens. B: Open weight Chinese that costs $0.50-1 per 1MM tokens. If you force American companies to spend 50-100x more than their competitors abroad here is what will
  • user avatar
    Eric Vishria
    Benchmark
    @ericvishria
    Jul 17
    This is an excellent post on how we go from controlled demos to real-world (ie, messy!), high success rate robotics. As many speculated, high quality pre-training and edge-case post-training, drives real-world generalizability. But holy shit making that theory work in
    user avatar
    Tony Zhao
    Sunday
    @tonyzzhao
    Jul 16
    Introducing ACT-2 Preview The first robotics model to unify broad generalization with high reliability. A single fine-tuning example can teach Memo a new behavior that generalizes. Zero shot, real unseen homes, 99% success rate.
    Image
    00:00
  • user avatar
    Eric Vishria
    Benchmark
    @ericvishria
    Jul 16
    10 minutes into my first meeting with @lqiao I knew I wanted to partner with her on Fireworks. The team and her still worked at Meta & Google, but the entrepreneur within was crystal clear. The idea has evolved tremendously from that initial pitch deck, but less than 4 years