Lists (18)
Alignment
DSP gang
efficient moe rl-tuning
LLM && Agents
LLM inference
LLM internal
How do you think?LLM PC
LLM pretraining
llm reasoning
LLM tuning
LLM4Sci
side projects --> startuplook-a-look
mamba(ssm)
multimodal
non-Trans LLMs
非Transformer结构的LLMsTriton && MLX && JAX
tts
text to speechworkflow
效率maxStars
ACE-Step: A Step Towards Music Generation Foundation Model
Latest Advances on Long Chain-of-Thought Reasoning
MiMo: Unlocking the Reasoning Potential of Language Model – From Pretraining to Posttraining
Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme
Accelerate LLM preference tuning via prefix sharing with a single line of code
LeanRL is a fork of CleanRL, where selected PyTorch scripts optimized for performance using compile and cudagraphs.
Fast Matrix Multiplications for Lookup Table-Quantized LLMs
📦 Repomix (formerly Repopack) is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Language Models (LLMs) o…
VPTQ, A Flexible and Extreme low-bit quantization algorithm
prime-rl is a codebase for decentralized RL training at scale
Code implementation of GPTQv2 (https://arxiv.org/abs/2504.02692)
Unleashing the Power of Reinforcement Learning for Math and Code Reasoners
GLM-4 series: Open Multilingual Multimodal Chat LMs | 开源多语言多模态对话模型
Technical report of Kimina-Prover Preview.
Democratizing Reinforcement Learning for LLMs
Kimi-VL: Mixture-of-Experts Vision-Language Model for Multimodal Reasoning, Long-Context Understanding, and Strong Agent Capabilities
Code for data-aware compression of DeepSeek models
Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory…
Official Implementation of QUAD: Quantization and Parameter-Efficient Tuning of LLM with Activation Decomposition
Official Implementation of LoRS: Efficient Low-Rank Adaptation for Sparse Large Language Model
g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains