🥧 Savoury implementation of the QUIC transport protocol and HTTP/3

LiteSpeed QUIC and HTTP/3 Library

HAProxy Load Balancer's development branch (mirror of git.haproxy.org)

QUIC and HTTP/3 implementation in Python

An implementation of the QUIC transport protocol.

Internet-Drafts that make up the base QUIC specification

An implementation of the IETF QUIC protocol

Cross-platform, C implementation of the IETF QUIC protocol, exposed to C, C++, C# and Rust.

A Survey and Benchmark of QUIC

Reasoning in LLMs: Papers and Resources, including Chain-of-Thought, OpenAI o1, and DeepSeek-R1 🍓

Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.

ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

AIFM: High-Performance, Application-Integrated Far Memory

TensorZero creates a feedback loop for optimizing LLM applications — turning production data into smarter, faster, and cheaper models.

The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

[ICLR2025] MagicPIG: LSH Sampling for Efficient LLM Generation

[ICLR 2025] LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs

Codes for the paper "∞Bench: Extending Long Context Evaluation Beyond 100K Tokens": https://arxiv.org/abs/2402.13718

InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)

Long context evaluation for large language models