1. X
  2. Benjamin Warner
Log inSign up
Benjamin Warner
693 posts
Benjamin Warner profile banner
user avatar

Benjamin Warner

@benjamin_warner
Research @SophontAI. Previously answerdotai. Vaccines save lives.
benjaminwarner.dev
Joined September 2011
513
Following
2,442
Followers
RepliesRepliesMediaMedia
  • Pinned
    user avatar
    Benjamin Warner
    @benjamin_warner
    Dec 19, 2024
    Today we released ModernBERT, the first encoder to reach SOTA on most common benchmarks across language understanding, retrieval, and code, while running twice as fast as DeBERTaV3 on short context and three times faster than NomicBERT & GTE on long context.
    Image
  • user avatar
    Benjamin Warner
    @benjamin_warner
    Aug 10
    Gotta love how Twitter's reliability is still terrible post purchase and conversion to X.
  • user avatar
    Benjamin Warner
    @benjamin_warner
    Aug 8
    As of codex 147, Luna is natively supported as a leaf-agent (it cannot spawn subagents)
    user avatar
    Marcos Hernanz
    @MarcosHernanz
    Aug 7
    If you're wondering how I spawn GPT 5.6 Luna subagents, here's the Codex config:
    Image
  • user avatar
    Benjamin Warner
    @benjamin_warner
    Aug 5
    What are the odds that there isn't a framework and that's why the WH won't release it or require open models to be tested against it?
    user avatar
    MTS
    @MTSlive
    Aug 4
    SITUATION UPDATE: The White House has exempted open models from its new framework to test frontier AI capabilities before release, per Axios.
  • user avatar
    Benjamin Warner
    @benjamin_warner
    Jul 30
    Congratulations to Liquid on a strong open-source encoder release. Looks like ModernBERT has some competition😄 My read from Liquid's benchmarks: - for multilingual try mmBERT, ModernBERT, & LFM-E - for english-only try ModernBERT & LFM-E
    user avatar
    Liquid AI
    @liquidai
    Jul 28
    Today we release LFM2.5-Encoder-230M and LFM2.5-Encoder-350M: bidirectional encoders that stay fast at long context, even on CPU. > LFM2.5-Encoder-230M: about 3.7x faster than ModernBERT-base on CPU at 8,192 tokens. Under 30s per forward pass, versus over a minute and a half. >
    Image
    00:00

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.
Advertisement
Advertisement