Skip navigation
Introducing HRM-Text.
0:00 / 5:31

Introducing HRM-Text.

Sapient Intelligence

@Sapient_Int6.2K followers

Visit on X
2,379 likesDownloadOriginal post

519,365 views

Introducing HRM-Text. An ultra-lean 1B-parameter reasoning language model designed to deliver strong general performance with a fraction of the data, compute, and infrastructure. Trained on just 40B structured tokens, HRM-Text achieves competitive performance while using ~1/1000 of the training data of comparable models. The kicker? The full model trains in roughly one day on a $1,000 budget. This opens the door to a new generation of AI that is powerful, accessible, and radically easier to adapt. Theories and research concepts once deemed too expensive to test are officially back in the game. Sapient Intelligence invites you to help us shape a new paradigm for general intelligence.

34 Comments

Comments from the original post on X (Twitter)

  • Sapient Intelligence's profile picture

    @Sapient_Int

    Download HRM-Text 🔗 Github: Hugging Face:

    275 likes

  • Flavius Burca's profile picture

    @flaviusburca

    Am I missing something ? HRM-Text is trained on a corpus where every benchmark's training set is present and often upsampled 10×. Of course you hit the same benchmark score with 150× less data when 150× more of your data is on-distribution for the eval.

    32 likes

  • AI_Explorer's profile picture

    @ai_explorer25

    Watched it twice. The $1k pretraining number is going to do more for independent research than any open-weights release this year

    27 likes

  • William Wale's profile picture

    @williawa

    Those are impressive results. Any reason you didn’t scale it up? It’s a bit strange to have a flagship that trains for a thousand dollars.

    23 likes

  • Flavius Burca's profile picture

    @flaviusburca

    The training mix (data_io) includes GSM8k-train ×10, MATH-train ×10, and FLAN (which contains HellaSwag/ARC/BoolQ/Winogrande train splits). Most of the 150× saving is the data mix, not the architecture. But I do have to applaude the ingenious H/L split :)

    12 likes

  • ρ:ɡeon's profile picture

    @pigeon__s

    those are the numbers for Qwen3 Qwen3.5s dataset size has not been disclosed and if you really believe in this AND its so cheap you should scale it up to at least 32B range and get llama.cpp compatability otherwise nobody can even use it or care about it

    11 likes

  • Crof AI's profile picture

    @aiCrof

    Very interesting! What about for bigger models larger than 1b?

    11 likes

  • DevHunterAI's profile picture

    @DevHunterYZ

    $1000 for a 1B parameter is excessive. I full pretrained my 1.5B RLM model with an RTX Pro 6000 for 13 hours. It cost less than $4(with Google Colab).

    8 likes

  • Knowix's profile picture

    @knowixbuilds

    looks promising, can't wait to try it out. I really need a small parameter model that I can run on my old PC

    8 likes

  • Brunão's profile picture

    @brunao_eth

    training a 1B model in a day for $1k is insane the barrier to AI research just dropped through the floor

    7 likes

  • Adam Ritter's profile picture

    @ritteradam

    Amazing result, this is where I don't understand why you didn't start a company instead of letting the LLM companies just use it to train much smarter models.

    7 likes

  • Everlier's profile picture

    @Everlier

    This release is incompatible with me going to sleep in time

    5 likes

  • Jules 𝕏's profile picture

    @JulesNetX

    Brilliant. Hierarchical dual-timescale reasoning in a 1B model that beats far larger ones on 40B tokens is exactly the efficiency breakthrough we need. This is how we get to real intelligence, not just scale.

    5 likes

  • himanshu's profile picture

    @himanshustwts

    awesome work, congrats!

    4 likes

  • Ziwen's profile picture

    @ziwenxu_

    wait it's only 1B and trained in a day?? That's wild

    4 likes

  • 𝐲𝐮𝐫𝐨's profile picture

    @yur_o_0

    "chain-of-thought" slander for 5 whole mins. great approach, moving from grid-based puzzling in HRM to implementing latent recursive reasoning in language modelling through HRM-text.

    4 likes

  • Sapient Intelligence's profile picture

    @Sapient_Int

    Paper:

    4 likes

  • BruzWJ's profile picture

    @BruzWJ

    one day and a thousand bucks is wild, but comparable models is doing a lot of work in that claim what evals are you running and what's the accuracy gap?

    3 likes

  • the.PM's profile picture

    @thePM_001

    That's super, so maybe we can use our new approach of cloning the LLM behavioural shadow here for only 10 USD and then inference it with CPU. Have to investigate.

    3 likes

  • Yuli Kay's profile picture

    @yulikay

    I’ve been using my personal agent for a while now So this could be a game-changer!

    3 likes