USASI
Model family

SmolLM

Family overview — summarizes releases; licenses and availability belong to each release.

Maintained by Hugging Face (Smol Models Research, HuggingFaceTB)13

SmolLM is Hugging Face's series of small language models. The first SmolLM release (July 2024) came in 135M, 360M, and 1.7B sizes; SmolLM2 kept those three sizes, with the 1.7B model trained on about 11 trillion tokens (paper February 2025); and SmolLM3 (July 2025) is a 3B model released as base and instruct checkpoints, with a dual-mode reasoning instruct model. The SmolVLM vision-language models are maintained in the same repository.45261

Last reviewedEntry updated Documented release Jul 16, 2024

Releases assessed

Each release is assessed on its own. This family has no family-wide openness label.
Releases in the SmolLM family
ReleaseWeightsTier (USASI rubric v0.1)LicenseReleased
SmolLM3 3BWeights: PublicModel-disclosure tier (USASI rubric v0.1): Fully openApache-2.0, Apache-2.0, Apache-2.0Jul 8, 2025

What it is useful for

Hugging Face describes the models as compact enough to run on-device, and publishes the training data mixtures, configurations, and code alongside the weights.12

Organization context

Provenance and derivatives

Trained from scratch by Hugging Face on public datasets. SmolLM used the SmolLM-Corpus (Cosmopedia v2, synthetic textbooks and stories generated with Mixtral; FineWeb-Edu; and Python-Edu, filtered from The Stack); SmolLM2 added the FineMath, Stack-Edu, and SmolTalk datasets; SmolLM3's pretraining mix includes FineWeb-Edu, DCLM, FineWeb-2, FineMath, and Stack-Edu among other public datasets.4578

U.S. eligibility

Project eligibility rests on documented governing or maintaining entities, not on contributors.

Eligible · basis: U.S. headquarters

The models are published by Hugging Face's HuggingFaceTB organization ("Hugging Face Smol Models Research") and documented in Hugging Face's GitHub repository. Hugging Face's terms of service identify Hugging Face, Inc., a Delaware corporation, as the provider of its services, and its privacy policy states that the company is located in the United States. See the hugging-face organization record for the dual-country assessment. Eligibility of each release is assessed on the release record.31910

Assessed Sep 29, 2026

Sources

  1. 1.
  2. 2.
    SmolLM family README (text/) (external site: raw.githubusercontent.com)

    Hugging Face · Documentation · accessed Sep 29, 2026

  3. 3.
    HuggingFaceTB (Hugging Face Smol Models Research) (external site: huggingface.co)

    Hugging Face · Official page · accessed Sep 29, 2026

  4. 4.
    SmolLM - blazingly fast and remarkably powerful (external site: huggingface.co)

    Hugging Face · Announcement · published Jul 16, 2024 · accessed Sep 29, 2026

  5. 5.
  6. 6.
    SmolLM3: smol, multilingual, long-context reasoner (external site: huggingface.co)

    Hugging Face · Announcement · published Jul 8, 2025 · accessed Sep 29, 2026

  7. 7.
  8. 8.
    The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale (arXiv 2406.17557) (external site: arxiv.org)

    Hugging Face (arXiv) · Paper · published Jun 25, 2024 · accessed Sep 29, 2026

  9. 9.
    Terms of Service (external site: huggingface.co)

    Hugging Face · Official page · accessed Sep 29, 2026

  10. 10.
    Hugging Face Privacy Policy (external site: huggingface.co)

    Hugging Face · Official page · accessed Sep 29, 2026

This listing is not an endorsement, a safety assessment, or a federal approval.

Support Us

Help keep USASI useful.

Optional. No USASI account required. Payment takes place on the linked provider’s website (Buy Me a Coffee).

About supporting this project