tiny-recursive-model

Tiny Recursive Model (TRM)

Implementation of Tiny Recursive Model (TRM), improvement to HRM from Sapient AI, by Alexia Jolicoeur-Martineau

Official repository is here

Interview with Alexia!

Paper review by bycloud

URM

Install

$ pip install tiny-recursive-model

Usage

import torch
from tiny_recursive_model import TinyRecursiveModel, MLPMixer1D, Trainer

trm = TinyRecursiveModel(
    dim = 16,
    num_tokens = 256,
    network = MLPMixer1D(
        dim = 16,
        depth = 2,
        seq_len = 256
    ),
)

# mock dataset

from torch.utils.data import Dataset
class MockDataset(Dataset):
    def __len__(self):
        return 16

    def __getitem__(self, idx):
        inp = torch.randint(0, 256, (256,))
        out = torch.randint(0, 256, (256,))
        return inp, out

mock_dataset = MockDataset()

# trainer

trainer = Trainer(
    trm,
    mock_dataset,
    epochs = 1,
    batch_size = 16,
    cpu = True
)

trainer()

# inference

pred_answer, exit_indices = trm.predict(
    torch.randint(0, 256, (1, 256)),
    max_deep_refinement_steps = 12,
    halt_prob_thres = 0.1
)

# save to collection of specialized networks for tool call

torch.save(trm.state_dict(), 'saved-trm.pt')

Citations

@misc{jolicoeurmartineau2025morerecursivereasoningtiny,
    title   = {Less is More: Recursive Reasoning with Tiny Networks}, 
    author  = {Alexia Jolicoeur-Martineau},
    year    = {2025},
    eprint  = {2510.04871},
    archivePrefix = {arXiv},
    primaryClass = {cs.LG},
    url     = {https://arxiv.org/abs/2510.04871}, 
}

Metadata

Release files for tiny-recursive-model 0.0.15

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for tiny-recursive-model 0.0.15
File	Size	Uploaded
tiny_recursive_model-0.0.15.tar.gz	10.0 kB	Dec 23, 2025	Details

Built distribution (wheel)

Table of built distributions (wheels) for tiny-recursive-model 0.0.15
File	Interpreter	ABI	Platform	Reset
tiny_recursive_model-0.0.15-py3-none-any.whl 11.0 kB Dec 23, 2025	Python 3	none	any	Details

Total release size: 21.0 kB

Release files / tiny_recursive_model-0.0.15.tar.gz

Download URL	tiny_recursive_model-0.0.15.tar.gz
Size	10.0 kB
Tags	Source
SHA-256 checksum How to use checksums	`ed3011a66be8dc365922f14c5dd2f5f31e4afc5f6191487e57c298fdd6e7cf97`
BLAKE2b-256 checksum How to use checksums	`554f82c0218b4fdd6911af55e6342de46d77bf951090319b98f765d5276eb445`
Upload date	Dec 23, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.9.25`

Release files / tiny_recursive_model-0.0.15-py3-none-any.whl

Download URL	tiny_recursive_model-0.0.15-py3-none-any.whl
Size	11.0 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`3099441e29f6f048b1b21c4bebf58ba4bc641039016ac686a0044bc22f4cc587`
BLAKE2b-256 checksum How to use checksums	`4d7bfc7f929a6482ad2fa541381738aae4d8bf9897e97e412433d3566b3dd8ee`
Upload date	Dec 23, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.9.25`