【NeurIPS 2024】Dense Connector for MLLMs

Inference script for Oasis 500M

AMES: Asymmetric and Memory-Efficient Similarity

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention)

Using FlexAttention to compute attention with different masking patterns

Code for "FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models (ACL 2024)"

Official repo for the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas"

Spec-Bench: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding (ACL 2024 Findings)

[COLM'24] Official Implementation of `Will the Real Linda Please Stand up...to Large Language Models? Examining the Representativeness Heuristic in LLMs`

[ECCV'24] Official Implementation of Autoregressive Visual Entity Recognizer.

Enhancing AI Software Engineering with Repository-level Code Graph

[NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

create your rotating proxy server with docker. self hosted rotating proxy service.

[ECCV 2024] official code for "Long-CLIP: Unlocking the Long-Text Capability of CLIP"

Helpful tools and examples for working with flex-attention

Diffusion Feedback Helps CLIP See Better

[ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference

Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.

Composable building blocks to build Llama Apps

VisionLLaMA: A Unified LLaMA Backbone for Vision Tasks

This is the repository for the GenIR survey.

[COLM 2024] TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding

This repo contains the source code for: Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs

Long Context Transfer from Language to Vision