Python 排行榜

PsyChat 是一款基于范例引导与Agentic RAG的智能 AI 心理咨询助手。基于高度匹配的Few-shot与学习笔记，应用理性情绪行为疗法（REBT）帮助用户识别并调整非理性信念，缓解负面情绪。

Local FAISS vector store as an MCP server – Agent Memory, drop-in local semantic search for Claude / Copilot / Agents.

Local LLM chat panel for ComfyUI — LM Studio & Ollama, multi-session, vision support, Guide Materials

Advanced automated image processing tool for selection, cropping, and standardization. (Helper for stable diffusion), now updated with GUI 🎉

[CVPR'2024] Adaptive Teacher-Student Collaboration for Text-Conditional Diffusion Models

Fine Tune Video Generator for pretrained Stable Diffusion And Disney Models

A simple UI tool to batch crop images to prepare datasets from images and videos.

Improved ReferenceLatent for ComfyUI — per-image strength, timestep gating, MediaPipe auto-masks, 1-4 image inputs. Works with Flux, Klein, Lumina, Wan, Hunyuan, Qwen-Image.

Panacea is a framework for building collaborative, intelligent multi agent AI systems. The framework provides a robust infrastructure for creating and managing multiple AI agents, and enables developers and organizations to build, deploy, and optimize AI agents that work well in dynamic, complex environments.

Code for LifelongMemory: Leveraging LLMs for Answering Queries in Long-form Egocentric Videos

VOXRAD is a voice transcription application for radiologists leveraging locally deployed ASR and LLM models.

[MLHC 2025] ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling

Turkish Speech Recognition using Facebook's Wav2vec 2.0 models

Fully local voice interface for Claude Code on Apple Silicon. Parakeet STT + Kokoro TTS + SmartTurn EOU + dual VAD.

Serve GLM-5.2 469B (REAP-pruned, NVFP4) across 3× NVIDIA DGX Spark with vLLM pipeline parallelism — 256K context, production-ready config and patches

Dual-engine (llama.cpp + vLLM) LLM benchmarking pipeline for GGUF & safetensors on NVIDIA GPUs — speed, quality, live dashboard, publishable cards.

Code for "BayesAdapter: Being Bayesian, Inexpensively and Robustly, via Bayeisan Fine-tuning"

PipelineLLM 是一个系统性的大语言模型（LLM）后训练学习项目，涵盖从监督微调（SFT）到偏好优化（DPO）、强化学习（RLHF/PPO/GRPO）再到持续学习（Continual Learning)的完整技术栈。

🔥 Tuning LLMs with 15+ Federated Fine-tuning Methods

Synthetic data generation to fuel AI models