OCNLI: 中文原版自然语言推理任务

PersonaPlex code.

整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.

My learning notes for ML SYS.

Tools for merging pretrained large language models.

[COLM 2025] Open-Qwen2VL: Compute-Efficient Pre-Training of Fully-Open Multimodal LLMs on Academic Resources

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

Latest Advances on System-2 Reasoning

A high-throughput and memory-efficient inference and serving engine for LLMs

SoulX-Podcast is an inference codebase by the Soul AI team for generating high-fidelity podcasts from text.

Model Compression Toolbox for Large Language Models and Diffusion Models

adds Sequence Parallelism into LLaMA-Factory

[NeurIPS 2025] NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation

A curated list of insanely awesome libraries, packages and resources for Quants (Quantitative Finance)

Democratizing Reinforcement Learning for LLMs

Get started with building Fullstack Agents using Gemini 2.5 and LangGraph

Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity. 🔔 Official updates only via twitter @Martin9…

Costrict - strict AI coder for enterprises, quality first, including AI Agent, AI CodeReview, AI Completion.

A symbolic notation designed for hyper-efficient communication and context management, primarily between Large Language Models (LLMs) and other AI systems.

This is a Python package to add tool calling capabilities to newly released LLMs on LangChain's ChatOpenAI, AzureAIChatCompletionsModel and ChatBedrockConverse classes ahead of time before LangChai…

No fortress, purely open ground. OpenManus is Coming.

Large Language Model based Multi-Agents: A Survey of Progress and Challenges (In IJCAI 2024)

SecGPT网络安全大模型

为学术搜索结果添加AI驱动的相关性评分，帮助研究人员快速识别相关论文

Minimal reproduction of DeepSeek R1-Zero

Fully open reproduction of DeepSeek-R1