mni-ml/transformer

A 12-million-parameter transformer language model implemented in Node.js and trained with the mni-ml/framework library on BPE-tokenized text such as TinyStories or YouTube-Commons transcripts.

Models

Visit website github.com

mni-ml/transformer is a small transformer language model built on top of the mni-ml/framework library. Training and text generation run in JavaScript on Node.js 22.18 or newer, while included Python scripts build the training dataset from the TinyStories corpus or YouTube-Commons video transcripts and train a byte-level BPE tokenizer.

The project includes separate CPU and GPU code paths for both training and sampling: the CPU path uses a matmul attention fallback that runs everywhere, and the GPU path uses CUDA or WebGPU through an optional native framework package. A default trained checkpoint and tokenizer are committed under out/, and other checkpoints stay local.

More in Models

See all 405

mini-AGI
github.com
A continual learning language model that trains from scratch on a single 8 GB VRAM GPU, assembling its own architecture and adapting its size to the hardware and data available rather than shipping as a fixed pretrained model.
- Models
Applied Inference Lab: Compliance Control Embedding Stack
github.com
An open-source contrastive-embedding pipeline that maps technical CVE vulnerability descriptions to formal cybersecurity compliance controls in NIST SP 800-53 and CMMC, published with its model generation, fine-tuning and evaluation code.
- Models
- Security
peekaboolean
github.com
A small vision-language model, fine-tuned with LoRA on a 500-million-parameter backbone, that answers typed multiple-choice, ordinal-score and yes/no questions about an image with calibrated probabilities, in about 400 milliseconds on a laptop.
- Models
- Inference

More in Models

mini-AGI

Applied Inference Lab: Compliance Control Embedding Stack

peekaboolean

Cookies