AI product Open source
Jevlike is an open-source research starter for training small models that choose among a variable list of text options. Given a context and an option list, it represents each option as a query, attends to the context tokens to produce an option-specific context vector, applies a shared dot product, and runs a softmax across the options to return probabilities in one pass rather than generating an answer token by token. It provides a trainable byte-level encoder and an optional frozen Hugging Face encoder, with command-line tools for generating synthetic data, training, evaluation, and prediction from JSONL records.
The repository also includes a visual option scorer used by Doom and chess controller examples. Training supports CPU, Apple MPS, and CUDA; the default encoder truncates contexts to 192 bytes and options to 32 bytes, and the project requires the complete option list at prediction time. Jevlike is an independent implementation rather than a copy of TypeSafe's commercial Jev model, and its code is released under the MIT License.
2 uses taken from transcripts — each links to the moment in the video.
Answers multiple-choice questions by scoring all options in one forward pass rather than generating text. It returns probabilities directly and is described as substantially faster than a small decoder for eight-option choices.
An open-source model that scores a changing list of text options in one pass, returning a probability for each option. It is intended for fast classification and selection tasks and can run on CPU, Apple MPS, or CUDA.
2 in the library.