close
Log inSign up
Omer Shlomovits profile banner
user avatar

Omer Shlomovits

@OmerShlomovits
Inference HW Acceleration; building @zroai_
Israel
zro.moonmath.ai
Joined February 2018
547
Following
3,154
Followers
RepliesMedia
  • user avatar
    Omer Shlomovits
    @OmerShlomovits
    Aug 6
    npm install -g @ moonmath-ai/zro zro login zro launch prime --model kimi-k3 zro launch prime --model deepseek-v4-flash-0731
    user avatar
    Zro
    @zroai_
    Aug 6
    one command to launch Prime Intellect and use open-weight frontier models optimized for agentic usage. most powerful model: > zro launch prime --model kimi-k3 fastest model: > zro launch prime --model deepseek-v4-flash-0731
    00:00
  • user avatar
    Omer Shlomovits
    @OmerShlomovits
    Aug 5
    Today we added @AMD MI325X as a new serving backend for @zroai_. Kimi K3 is the first model running on it, powered by @sgl_project. We believe this is the first production support for the official Kimi K3 weights on CDNA3. With 256 GB of HBM per GPU, the full model fits on a
  • user avatar
    Omer Shlomovits
    @OmerShlomovits
    Aug 5
    Those are… unusually low prices 🧐 All three citations point to the same URL, and the quoted rates look more like heavily discounted or long-term reserved pricing than on-demand. Normalize using actual on-demand market prices, and the picture changes completely.
    user avatar
    wafer
    @wafer_ai
    Aug 1
    🚨 BREAKING: these engineers figured out how to serve Kimi K3 on @AMD MI355X at 952 tok/s/node and 118 tok/s single stream! this crushes B200 by 3.8x in aggregate throughput/node and 1.3x in single stream decode + beats B300 on performance per dollar (48 vs 33 tok/s/$) See how
  • user avatar
    Omer Shlomovits
    @OmerShlomovits
    Aug 5
    We are here to serve
    user avatar
    Steve 🇺🇸
    @0x3_dev
    Aug 5
    One of the best, cheapest, and fastest providers for the new DeepSeek Flash v4 is @zroai_. I’ve been testing them over the last 24 hours, and I haven’t hit a single rate limit, error, or other issue. TPS is also insanely fast. I’m not affiliated with them in any way — just
  • user avatar
    Omer Shlomovits
    @OmerShlomovits
    Jul 31
    The updated dsv4-Flash-0731 feels unfairly good :) We added day-@zroai_ support serving it from our EU infra. And, just like with Kimi K3, we cut the price 🤠
    user avatar
    DeepSeek
    @deepseek_ai
    Jul 31
    🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta! 🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇 🔷 The official V4-Flash now natively supports the