Recipes to scale inference-time compute of open models

Rich is a Python library for rich text and beautiful formatting in the terminal.

Joplin markdown plugin for custom containers (Updated for Joplin 3.1.24)

End-to-end Generative Optimization for AI Agents

A recipe for online RLHF and online iterative DPO.

Beyond Accuracy: Behavioral Testing of NLP models with CheckList

Node and browser JavaScript library to run simulations. Supports System Dynamics modeling, Differential Equation mathematical modeling, and Agent Based Modeling.

Building blocks for rapid development of GenAI applications

Tabular Deep Learning Library for PyTorch

Provides a language server for grammars based on Lark

TARGET is a benchmark for evaluating Table Retrieval for Generative Tasks such as Fact Verification and Text-to-SQL

Simple Website to share notes

Monte Carlo tree search in JAX

Probabilistic programming with HuggingFace language models

Composable inference algorithms with LLMs and programmable logic

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

📰 Must-read papers and blogs on Speculative Decoding ⚡️

[ICML 2024] LLMCompiler: An LLM Compiler for Parallel Function Calling

maximal update parametrization (µP)

A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).

Evaluation suite for LLMs

Modeling, training, eval, and inference code for OLMo

🦄 Record your terminal and generate animated gif images or share a web player

Code for Palu: Compressing KV-Cache with Low-Rank Projection

Efficient and general syntactical decoding for Large Language Models

Formatron empowers everyone to control the format of language models' output with minimal overhead.

Syntax highlighting for ABNF/BNF/EBNF, Yacc, and other language-related languages.

A high-performance constrained decoding engine based on context free grammar in Rust