Log inSign up
0xNeoArch profile banner
user avatar

0xNeoArch

@0xNeoArch
Software engineer • AI orchestration • Agentic AI SDLC frameworks • Local AI & hardware • Infra • DevOps Sharing what works
Joined September 2024
4,495
Following
294
Followers
RepliesMedia
  • Pinned
    user avatar
    0xNeoArch
    @0xNeoArch
    Jul 17
    Getting ready to download Kimi K3 weights 🚀 I used to mine $XCH and still have enterprise-grade infra with NetApp storage and 72x20TB drives - all the expensive cabling and full UPS setup. 1.44 PB raw capacity today! I got the infra and NetApps for additional 168 drives
  • user avatar
    0xNeoArch
    @0xNeoArch
    5h
    LTX 2.5 weights drop in a few hours. Open. Free. Day one. While the big labs hand you a waitlist and an API bill, these folks just post a download link. Set your reminder. This one's going to be everywhere by tonight. ltx.io/2-5-open-weigh…
  • user avatar
    0xNeoArch
    @0xNeoArch
    17h
    Post like these make me go all in and buy more GPUs 😇 I can imagine in a few years speed like this will be the new normal.
    user avatar
    am.will
    @LLMJunky
    19h
    anything under deepseek v4 flash basically doesn't exist to me. with <200GB of vram, i'm getting ~800K context @ 200 tokens per second. It's not quite Cerebras fast, but damn. It is a joy to use, and its basically somewhere in the opus 4.7/4.8 range. It's incredibly good at
    00:00