SparkJiao

dspy

DSPy: The framework for programming—not prompting—foundation models

gpt-fast

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

gemma_pytorch

The official PyTorch implementation of Google's Gemma models

OLMo

Modeling, training, eval, and inference code for OLMo

mergekit

Tools for merging pretrained large language models.

Qwen1.5

Qwen1.5 is the improved version of Qwen, the large language model series developed by Qwen team, Alibaba Cloud.

Pearl

A Production-ready Reinforcement Learning AI Agent Library brought by the Applied Reinforcement Learning team at Meta.

lightllm

LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalability, and high-speed performance.

open-instruct

dlrover

DLRover: An Automatic Distributed Deep Learning System

DeepSeek-MoE

nanotron

Minimalistic large language model 3D-parallelism training

megablocks

llama-moe

⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training

alpaca_farm

A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

EAGLE

[ICML'24] EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty

flash-linear-attention

Efficient implementations of state-of-the-art linear attention models in Pytorch and Triton

lumos

Code and data for "Lumos: Learning Agents with Unified Data, Modular Design, and Open-Source LLMs"

CoLLiE

Collaborative Training of Large Language Models in an Efficient Way

Awesome-Reasoning-Foundation-Models

✨✨Latest Papers and Benchmarks in Reasoning with Foundation Models

ChunkLlama

[ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"

zero-bubble-pipeline-parallelism

Zero Bubble Pipeline Parallelism

HallusionBench

[CVPR'24] HallusionBench: You See What You Think? Or You Think What You See? An Image-Context Reasoning Benchmark Challenging for GPT-4V(ision), LLaVA-1.5, and Other Multi-modality Models

Q-Instruct

②[CVPR 2024] Low-level visual instruction tuning, with a 200K dataset and a model zoo for fine-tuned checkpoints.

RefGPT

RLCD

Reproduction of "RLCD Reinforcement Learning from Contrast Distillation for Language Model Alignment

SimulateBench

GPT as Human

SeaEval

NAACL 2024: SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoning

dpo-trajectory-reasoning

Source code for the paper "Learning Planning-based Reasoning with Trajectory Collection and Process Rewards Synthesizing".

RLMEC

The official repository of "Improving Large Language Models via Fine-grained Reinforcement Learning with Minimum Editing Constraint"