Skip navigation
introducing rd-signal-2: a frontier classification model that is 1600x cheaper than GPT 5.6 Sol.
0:00 / 0:49

introducing rd-signal-2: a frontier classification model that is 1600x cheaper than GPT 5.6 Sol.

ben hylak

@benhylak54K followers

Visit on X
995 likesDownloadOriginal post

164,473 views

introducing rd-signal-2: a frontier classification model that is 1600x cheaper than GPT 5.6 Sol. free to try in Raindrop, and available via a new API for training/hosting custom classifiers with Zero Data Retention.

57 Comments

Comments from the original post on X (Twitter)

  • ben hylak's profile picture

    @benhylak

    read more at: rd-signal-2 balances precision/recall at a price that is still 280x lower than GPT-5.6 Luna xhigh. it powers signals and real-time issue detection in raindrop, allowing our customers to detect and accurately track sparse behaviors.

    56 likes

  • ben hylak's profile picture

    @benhylak

    most LLM observability providers require customers to sample a very small % of traffic (and pay for the judges!) but agent failures are often rare. it's common for a behavioral anomaly to be < 1% of traffic. sampling can completely hide the problem.

    35 likes

  • ben hylak's profile picture

    @benhylak

    you can say "the agent saying it transferred to a human, but didn't call the tool" or "the user complaining that web search results were stale" @raindrop_ai signals will write the code, assemble the context, train a custom classification head for your task, and self-improve.

    32 likes

  • ben hylak's profile picture

    @benhylak

    rd-signal-2 also powers @raindrop_ai's real-time issue detection. many emerging failures are detected in under 5 minutes. because they share the same foundation, any raindrop issue can be instantly converted to a signal, compared in experiments, and more.

    25 likes

  • ben hylak's profile picture

    @benhylak

    rd-signal-2 already processes around 20 billion traces a month. members of the @raindrop_ai team: - invented fraud transformer models at robinhood - built the first recommender system at pinterest - have built agents that have millions of customers - and more dm me to join

    30 likes

  • ben hylak's profile picture

    @benhylak

    stop paying for judges. starting using signals. try the agent observability platform trusted by vercel, speak, and the fortune 100. read more at

    38 likes

  • eric provencher's profile picture

    @pvncher

    @raindrop_ai Damn your motion design is on point! Looks awesome

    7 likes

  • ben hylak's profile picture

    @benhylak

    @raindrop_ai thank you!

    1 like

  • gian's profile picture

    @giansegato

    @raindrop_ai sweeet, great job! adaptive online observability is basically table stakes at this point

    6 likes

  • ben hylak's profile picture

    @benhylak

    @raindrop_ai we agree.

    1 like

  • Soleio's profile picture

    @soleio

    @raindrop_ai “Stop paying for judges. Start using signals.” :chefs_kiss:

    5 likes

  • Eduard Faus Gil's profile picture

    @eduardfausgil

    @raindrop_ai Typical raindrop W so glad I told @orchid_hq to use you guys

    5 likes

  • Logan Ford's profile picture

    @lhford0

    @raindrop_ai very excited to try this out!

    4 likes

  • Dev's profile picture

    @devparagiri

    @raindrop_ai hey @alialobai1 @hegargarcia 👀

    4 likes

  • ben hylak's profile picture

    @benhylak

    @raindrop_ai @alialobai1 @hegargarcia would love your feedback :)

    1 like

  • bani's profile picture

    @banisgh

    @raindrop_ai huge

    4 likes

  • ben hylak's profile picture

    @benhylak

    @raindrop_ai thanks bani! :)

  • Nizzy's profile picture

    @nizzyabi

    @raindrop_ai woahhh

    4 likes

  • ben hylak's profile picture

    @benhylak

    @raindrop_ai give it a try!

    2 likes

  • Sam Hogan 🇺🇸's profile picture

    @samhogan

    @raindrop_ai very cool

    3 likes