Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.

Textbook on reinforcement learning from human feedback

A project to scrape, tabulate, and display job data from the O*NET website, and possibly other websites. Non-commercial.

slime is an LLM post-training framework for RL Scaling.

Miles is an enterprise-facing reinforcement learning framework for large-scale MoE post-training and production workloads, forked from and co-evolving with slime.

KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

A benchmark for LLMs on complicated tasks in the terminal

Modular Multi-Agent System for Scientific Research Assistance

OfficeBench: Benchmarking Language Agents across Multiple Applications for Office Automation

"Towards Improving Document Understanding: An Exploration on Text-Grounding via MLLMs" 2023

SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images (AAAI2023)

Contexts Optical Compression

Official github repo for C-Eval, a Chinese evaluation suite for foundation models [NeurIPS 2023]

DocBench: A Benchmark for Evaluating LLM-based Document Reading Systems

Parse LaTeX math expressions

[Neurips 2025] R-KV: Redundancy-aware KV Cache Compression for Reasoning Models

[EMNLP 2024 (main)] Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters

APOLLO: SGD-like Memory, AdamW-level Performance; MLSys'25 Oustanding Paper Honorable Mention

A high-throughput and memory-efficient inference and serving engine for LLMs

GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM

More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache Compression

Awesome-LLM-KV-Cache: A curated list of 📙Awesome LLM KV Cache Papers with Codes.

Unified KV Cache Compression Methods for Auto-Regressive Models

The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.

🧮 A collection of resources to learn mathematics for machine learning

Code for our EMNLP 2023 Paper: "LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models"