Python 排行榜

A python library to interact with AI-Horde's free generative AI APIs

[ACL 2024] An Easy-to-use Hallucination Detection Framework for LLMs.

🎬 AI-powered localhost subtitle generator for hearing-impaired users. Automatic speech recognition using OpenAI Whisper + Intel OpenVINO. Supports 99 languages, torrent downloads, and online video playback.

Windows AI 语音输入🎙 — 按快捷键说话即输入，支持润色。摆脱打字限制，实现无拘束、高效率的表达。

A dinamic virtual assistent made with Python, you can easily add more voice commands without any code

Native Windows vLLM 0.27.1 wheels: Python 3.13, PyTorch 2.13 + CUDA 13.0, SM 7.5-12.0 for RTX 20/30/40/50, OpenAI-compatible serving, FlashAttention/Rust, 10 KV formats, Multi-TurboQuant, and experimental CPU/RAM/NVMe prompt-KV offload - no WSL or Docker.

Evano Studio is an open-source local AI desktop app for agents, automation, knowledge, and productivity.

Understand how CLI AI agents work and use your own local CLI AI! Easy to understand, written in python, built on Ollama/OpenAI compatible APIs.

Ollama Shell Helper (osh) : English to Unix-like Shell Commands translation using Local LLMs with Ollama

TraceTree - Runtime behavioral analysis tool that maps the process cascade of suspicious packages into a directed tree, catching supply chain attacks that install-time scanners miss.

Self-hosted chat client for Ollama, OpenAI, and Anthropic. Web UI and terminal CLI, with web search and a browser-automation agent.

Language Model Fine-tuning for Moby Dick

Pure RL to post-train base models for social reasoning capabilities. Lightweight replication of DeepSeek-R1-Zero with Social IQa dataset.

[AAAI 2026] D²PPO: Diffusion Policy Policy Optimization with Dispersive Loss.

See also APPL: https://github.com/appl-team/appl that improves this project. A Python package for writing Language Models prompts in a nested way.

PyCrust is a CLI tool that converts Python code to Rust using OpenAI's GPT-3.5 language model for code generation. It provides an easy-to-use interface for converting Python code to Rust code, making it easier for Python developers to leverage Rust's performance benefits.

Serving 4-bit Qwen3.8-27B on a single DGX Spark (GB10): 75 tok/s single-stream, 246 tok/s aggregate at 8-way concurrency. NVFP4 vs MixedInt4-AutoRound vs the FP8 baseline, measured on one harness — including why the quantization advantage collapses to +0.2% by c16.

Code for "Finetuning Pretrained Transformers into Variational Autoencoders"

An open source community implementation of the model from "DIFFERENTIAL TRANSFORMER" paper by Microsoft.

Implementation of a modular, high-performance, and simplistic mamba for high-speed applications