CARD · Card record

MiniMind: learn how a small language model is trained

MiniMind provides a roughly 64M-parameter model and training tutorials. Its “two hours” refers to one supervised fine-tuning epoch on a single RTX 3090.

01
THE STORY · The original

Read the explanation and its limits.

For learning the training process: start with examples and hardware requirements.

  • Originalattention signalhttps://github.com/jingyaogong/minimind

Why it is worth understanding

Use a small exercise to connect data, settings and outputs. Choose it for a clear learning goal, rather than its chart position.

Conditions and limitations

The timing is a maintainer report under specific conditions, not the full training workflow. We have not reproduced it; Star counts do not establish results or production suitability.

Important additions and corrections

Dates below mark changes to our coverage.

  • Addition ·

    Clarified the training conditions: “two hours” means one supervised fine-tuning epoch on an RTX 3090, not the full training process.

Reading progress is saved in this browser. Save or add a note →

02
THE THREAD · Full timeline

1 event, every row traceable.

A card accumulates records as its story develops: new corroborators, market contrasts, material changes at the original — each dated and linked to its archive day.

Card timeline
  1. RecordGitHub weekly Trending ranked #5, recording 3,649 stars for the week.

03
SOURCES · Sources and evidence

Check each link's purpose and limits.

Sources and evidence · 2 links

Each link has a purpose and scope. Link counts do not establish independent confirmation; direct support applies only to the statements specified below.

Found an error or a problem? Send feedback ↗

04
EDITORIAL REVIEW

Review of this card's wording and evidence.

Editorial review · Evidence reviewed

· Beijing time (UTC+8)

Read source materials to clarify purpose and limits. Discovery and event dates are retained; editorial review is not an independent experiment.

This dates our review of the card's wording and evidence, not a project release or product update.

Review sources and full receiptFull receipt: before, after and review evidence (JSON)