An Analytical Evaluation Board of Multi-turn LLM Agents [NeurIPS 2024 Oral]

AgentTrace is a lightweight observability library to trace and evaluate agentic systems.

A very simple GRPO implement for reproducing r1-like LLM thinking.

Recipes to scale inference-time compute of open models

SOTA Open Source TTS

ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search (NeurIPS 2024)

The first AI agent that builds permissionless integrations through reverse engineering platforms' internal APIs.

Open-source, self-hosted note-taking tool built for quick capture. Markdown-native, lightweight, and fully yours.

MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

The Open Source AI-Powered Code Editor. A fork of VSCode and Continue and PearAI

An app and set of methodologies designed to evaluate the performance of various Large Language Models (LLMs) on the text-to-SQL task. Our goal is to offer a standardized way to measure how well the…

Deita: Data-Efficient Instruction Tuning for Alignment [ICLR2024]

800,000 step-level correctness labels on LLM solutions to MATH problems

Implementation of the ICML 2024 paper "Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learning" presented by Zhiheng Xi et al.

"Self-Supervised Contrastive Learning for Long-term Forecasting", accepted at International Conference on Learning Representations (ICLR) 2024

container image to single executable compiler

Minimalist ML framework for Rust

[NeurIPS 2024 D&B] GTA: A Benchmark for General Tool Agents & [arXiv 2026] GTA-2

The context API to search, scrape, and interact with the web at scale. 🔥

llama3.np is a pure NumPy implementation for Llama 3 model.

Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

Simplify your onnx model

A toy large model for recommender system based on LLaMA2/SASRec/Meta's generative recommenders. Besides, note and experiments of official implementation for Meta's generative recommenders.

Repository hosting code for "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152).

Code for the CVPR 2024 paper highlight and demo "PIGEON: Predicting Image Geolocations".

Fast, flexible LLM inference