microgpt

A from-scratch, dependency-free GPT implementation in pure NumPy: byte-level tokenizer, multi-head self-attention, Adam trainer, and top-k sampler.

prabeakarve/microgpt

View on GitHub ↗

A from-scratch, dependency-free GPT implementation in pure NumPy: byte-level tokenizer, multi-head self-attention, Adam trainer, and top-k sampler.

microgpt ranks 110th of 113 in Pretraining & Training, last commit 16 days ago.

Star trajectory

14 days tracked
2026-11-0851 stars today2026-11-22
Momentum

- in 30d, below the Pretraining & Training median of +0.3%.

Maintenance

Last commit 16 days ago, fresher than the median of 103d.

How it compares in Pretraining & Training

microgpt
51
stars
- 30d
commit 16 days ago
Pretraining & Training median
342
stars
+0.3% 30d
commit ~103d ago
Leader · maxtext
2.5k
stars
+1.0% 30d
commit today

Ranked 110th of 113 by stars.

Ranked by shared tags, then keywords, then stars
  1. :sunrise: A comprehensive, cutting-edge open-source framework for next-generation Transformers/LLM, with…

    Committed 2d ago · Python · MIT

    Shares Training Toolkits, Pretraining & Training

  2. Projects from basic algorithms to MARL. Implements MADDPG,MATD3,MA/HAPPO in Predator-Prey pursuit games with…

    +9 (+2.0%) in 30d · Committed 103d ago · Jupyter Notebook · MIT

    Shares Training Toolkits, Pretraining & Training

  3. InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the…

    -1 (-0.2%) in 30d · Committed 458d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits

  4. Cevahir Zihin: a mind that learns by living, not a Transformer. It keeps knowledge as evidence on ties…

    +52 (+20%) in 30d · Committed today · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits

  5. Official Repo of "RobustFlow: Towards Robust Agentic Workflow Generation"

    +1 (+0.4%) in 30d · Committed 398d ago · Python

    Shares Pretraining & Training, Training Toolkits

  6. Autonomous Qwen3-VL training-code research on the official DocVQA benchmark. main: NVIDIA multi-GPU, mlx…

    +0 (+0.0%) in 30d · Committed 22d ago · Python · MIT

    Shares Pretraining & Training, Training Toolkits

  7. a LLM cookbook, for building your own from scratch, all the way from gathering data to training a model

    +2 (+1.1%) in 30d · Committed 879d ago · Jupyter Notebook · MIT

    Shares Pretraining & Training, Training Toolkits

  8. [NeurIPS'25] Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning

    +1 (+0.6%) in 30d · Committed 326d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits

  9. Typed-decision models (noul / choice / score) trained by a self-improving loop of AI agents — checkpoints, the…

    +19 (+16%) in 30d · Committed yesterday · HTML · MIT

    Shares Training Toolkits, Pretraining & Training

  10. Frontier-level agent capabilities with nothing more than high-quality data and a simple training recipe: 2 SFT…

    Committed 10d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits

Language
Python
License
MIT License
Contributors
1
Forks
0
Watchers
0
Open issues
0
Releases
1
Created
2026-11-05
Stars this week
+0

You've seen where microgpt stands today.

Get the moves before everyone else: the 5 Pretraining & Training repos climbing fastest, every Tuesday.

Free weekly digest. Double opt-in, unsubscribe anytime.

Maintain microgpt?

Claim this page to add your own call-to-action.

Claim this page