SWE-bench: Can Language Models Resolve Real-world Github Issues?

UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)

Presentation Slides for Developers

Elevate your AI research writing, no more tedious polishing ✨

Repository for the paper "Large Language Model-Based Agents for Software Engineering: A Survey". Keep updating.

AIInfra（AI 基础设施）指AI系统从底层芯片等硬件，到上层软件栈支持AI大模型训练和推理。

LaTeX template for USTC thesis

Accelerating Long Context LLM Inference with Accuracy-Preserving Context Optimization in SGLang, vLLM, llama.cpp, OpenClaw, RAG, and Agentic AI.

Kimina-Prover RL pipeline

Kimina Lean server (+ client SDK)

Major CS conference publication stats (including accepted and submitted) by year.

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

Serverless LLM Serving for Everyone.

Official repository for the EMNLP 2025 paper "Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistency".

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

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

[ICLR 2025🔥] SVD-LLM & [NAACL 2025🔥] SVD-LLM V2

A PyTorch native platform for training generative AI models

An interference-aware scheduler for fine-grained GPU sharing

NVIDIA Linux open GPU kernel module source

Easy and Efficient dLLM Fine-Tuning

LM engine is a library for pretraining/finetuning LLMs

LaTeX Template for Statement of Purpose (SoP)

dInfer: An Efficient Inference Framework for Diffusion Language Models

[SIGMOD 2025] PQCache: Product Quantization-based KVCache for Long Context LLM Inference

Pie: Programmable LLM Serving

An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

Kernels, of the mega variety :)