Python 排行榜

Layered prefill changes the scheduling axis from tokens to layers and removes redundant MoE weight reloads while keeping decode stall free. The result is lower TTFT, lower end-to-end latency, and lower energy per token without hurting TBT stability.

Network-faithful simulation of LLM serving and training deployments

[AAAI 2025 oral] Attribution Analysis Meets Model Editing: Advancing Knowledge Correction in Vision Language Models with VisEdit

A Scheduler for Batched LLM Inference

A comprehensive GUI wrapper application for Apple's Foundation Models Adapter Training Toolkit, providing an intuitive interface for training LoRA adapters for Apple's on-device foundation models.

Fast, config-driven SDXL LoRA & DreamBooth fine-tuning — train from a single YAML file. QLoRA + torch.compile.

Deferred Continuous Batching in Resource-Efficient Large Language Model Serving (EuroMLSys 2024)

LLM-adaptive embeddings (Zero-shot / LoRA) with Generative Topic Modeling & Agent-based workflow for social science text mining

Implementation of TSDS: Data Selection for Task-Specific Model Finetuning. An optimal-transport framework for selecting domain-specific and task-specific training data to improve LLM finetuning and instruction tuning.

A feature-rich web application for automated news scraping and summarization. It allows users to enter article URLs, fetches the full content, and generates concise summaries. The system supports both local inference with custom models and remote deployment via FastAPI or Streamlit interfaces.

(WACV'24) Kaizen: Practical self-supervised continual learning with continual fine-tuning

KAREN: Unifying Hatespeech Detection and Benchmarking

Running Stable diffusion from a mobile phone using Termux

A Python package to make Stable Diffusion Image Generation ridiculously easy

Toolchain for creating custom datasets and training Stable Diffusion (1.x, 2.x, XL) models and LoRAs

cli tool to search for models/datasets on huggingface and export them easily

Integrated Semantic and Phonetic Post-correction for Chinese Speech Recognition

A starter kit for evaluating benchmarks on the 🤗 Hub

Collection of recent advanced RAG techniques.

Advanced PDF/Document Translator with interactive comparison. Built on IBM Docling.