Models

An open-source reinforcement-learning framework that trains multi-step agents on real-world tasks using GRPO, so a model learns from the outcome of its own actions.

• Agent frameworks
• Models

A family of small, on-device large language models for edge and resource-constrained deployment, with the newest 2-billion-parameter release reporting state-of-the-art results for its size class.

• Models

A hands-on tutorial and codebase that implements a transformer language model from scratch in plain PyTorch, walking from raw text through pretraining to a fine-tuned, reasoning-style chat model.

• Models
• Tutorials

A declarative deep-learning framework, hosted by the Linux Foundation, for training, fine-tuning and deploying large language models and other AI models from a YAML configuration file instead of custom code.

• Models

A one-stop retrieval toolkit for search and retrieval-augmented generation, providing a family of open embedding, reranking and multimodal-search models known as BGE.

• RAG
• Models

An open-source, bilingual Chinese-English conversational language model with tool-calling support, developed by Fudan University, including model weights, dialogue training data, and deployment and fine-tuning code.

• Models

A free, open-source framework for fine-tuning large language models, with fast-moving support for new model families and training techniques such as mixture-of-experts LoRA and multi-dimensional parallelism.

• Models

A tiny, 8-to-29-megabyte foundation model for tool calling, structured data extraction and text embedding, designed to run directly on phones, wearables, robots, cars and microcontrollers.

• Models

A large-language-model development kit built on the PaddlePaddle framework, supporting efficient training across hardware types, lossless compression and high-performance inference.

• Models
• Inference

A Transformers-style training library built on the PaddlePaddle deep-learning framework, giving it model interfaces comparable to Hugging Face Transformers for training large language and vision-language models.

• Models

A library of more than 20 large language models implemented from scratch in plain PyTorch, with recipes to pretrain, fine-tune and deploy them at any scale from one GPU to a thousand.

• Models

A framework that automates high-value research-and-development work in AI, letting AI agents drive the model and data side of AI research, from ideation through implementation to evaluation.

• Agent frameworks
• Models

An RNN architecture that reaches transformer-level language-model performance while training in parallel like a transformer, with linear time, constant memory, no key-value cache and unlimited context length.

• Models

A framework for building local, private, enterprise retrieval-augmented generation applications, pairing a catalog of 300+ small, optimized language models with document parsing and knowledge-base pipelines.

• RAG
• Models

An independent, open-source research reimplementation of the Claude Mythos model architecture, a recurrent-depth transformer with a looped block and a sparse mixture-of-experts feed-forward network.

• Models

An application for building datasets to fine-tune, retrieval-augment and evaluate large language models, with document parsing, text splitting, question generation and data cleaning built in.

• Models
• Data engineering

A large model and multimodal model fine-tuning and deployment framework from the ModelScope community, supporting training, inference, evaluation, quantization and deployment for more than 600 text and 400 multimodal models.

• Models
• Inference

An open-source AI Development Environment that turns a code editor into a place to build, debug and deploy LLM and machine learning pipelines, running on a multithreaded C++ engine with any model, tool or framework.

• Models
• Data engineering

A Rust tensor library and deep learning framework that unifies training and inference in one codebase, JIT-compiling tensor operations with automatic kernel fusion across CUDA, ROCm, Metal, Vulkan and WebGPU.

• Models
• Inference

A lightweight deep learning inference and training framework, used in production across more than 30 Alibaba apps, with companion projects for running large language and stable diffusion models locally on phones, PCs and IoT devices.

• Models
• Inference

A GPU-optimized library for training transformer language models at scale, providing both a ready-to-run reference training setup and composable building blocks for custom large-model training pipelines.

• Models

A one-stop tool for creating a personal AI chat persona from exported chat history, fine-tuning a language model on the logs to match an individual's writing style and connecting the result to a chatbot.

• Models

A lightweight reinforcement learning framework for training AI agents using their existing, unmodified harnesses, by proxying model calls and turning the resulting interactions into training data.

• Agent frameworks
• Models

Open Chinese LLaMA and instruction-tuned Alpaca language models that extend the original LLaMA vocabulary for Chinese, further pretrained and fine-tuned on Chinese text and instructions, for local CPU or GPU deployment.

• Models