p-e-w/heretic
Fully automatic censorship removal for language models
A star delta has been computed for 61,465 of 62,212 tracked repositories in this period, and only those can be ranked in it.
A repository missing here has not been measured far enough back yet. That does not mean it did not move.
Star counts measured between 16:22 and 08:09 (BKK).
Fully automatic censorship removal for language models
A high-throughput and memory-efficient inference and serving engine for LLMs
SGLang is a high-performance serving framework for large language models and multimodal models.
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
Faster Whisper transcription with CTranslate2
A framework for efficient model inference with omni-modality models
Port of OpenAI's Whisper model in C/C++
An open, programmable decision layer for models and compute.
FreeToken brings datacenter-scale model serving to your desktop. Run massive models locally, fast and efficiently.
MNN: A blazing-fast, lightweight inference engine battle-tested by Alibaba, powering high-performance on-device LLMs and Edge AI.
Community maintained hardware plugin for vLLM on Huawei Ascend
SOTA Open Source TTS
Translate manga/image 一键翻译各类图片内文字 https://cotrans.touhou.ai/ (no longer working)
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
免選字的注音輸入法：在本機用 Transformer 看前文選字。支援 Linux（Fcitx5／IBus）、macOS、Windows。
PyTorch native quantization for training and inference
QuantMind（量化大脑）开源版是一款面向个人开发者与投研团队的 AI 原生多市场量化交易平台。深度集成微软 Qlib、RD-Agent 因子演化与 QuantBot全能工作台，提供从 300+ 维因子挖掘、13 种机器学习与深度学习模型工场、Qlib 高性能回测、截面批量推理、7x24 实时舆情情绪分析，到通达信深度联动（板块推送/预警雷达/闪电下单）与实盘模拟交易的完整闭环。支持 A股、港股、美股、期货及区块链。可免费商用，系统采用 Docker Compose 一键私有化部署，数据与模型完全本地化，保障策略隐私，功能零门槛无限制。
Scalable and user friendly neural :brain: forecasting algorithms.
SFML is the first fully local AI solar forecast for Home Assistant, powered by a local Attention Transformer. No external AI — such as ChatGPT, Gemini, or Grok — required. Runs entirely on your device for complete privacy.
A language model trained from scratch and running offline on an ESP32-S3 with a touchscreen chat interface.
Quantization, kernels, runtime and inference engine for mobiles, wearables, smart home and robots.
A typescript-go toolchain, with evidence graph
A Laravel Fractal package for building API responses, giving you the power of Fractal with Laravel's elegancy.
Zig INferenCe Engine — Local LLM inference on AMD GPUs and Apple Silicon
Pre-trained Neural Network models in Axon (+ 🤗 Models integration)
Unified LLM gateway with weighted load balancing, observability & cost tracking. 统一的 LLM 网关，提供权重负载均衡、可观测性与费用追踪。
Microsoft Text-to-Speech API sample code in several languages, part of Cognitive Services.
cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.
Graphormer is a general-purpose deep learning backbone for molecular modeling.
Open-source deep learning framework for Java. Designed with speed and lightweight in mind.
Showing 30 of 34 repositories