Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.
Implement a reasoning LLM in PyTorch from scratch, step by step
MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model.
[Neurips 2025] R-KV: Redundancy-aware KV Cache Compression for Reasoning Models
Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B
DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models
A skill to kill from-scratch coding — Claude checks real arXiv prior art before it designs a new architecture.
Official implementation of the NeurIPS 2025 paper "Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space"
[ACL 2026 Oral] "LightReasoner: Can Small Language Models Teach Large Language Models Reasoning?"
[NeurIPS 2025] Simple extension on vLLM to help you speed up reasoning model without training.
Official repository for EXAONE Deep built by LG AI Research
MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphs
This repository hosts the instructions and workshop materials for Lab 333 - Evaluate Reasoning Models for Your Generative AI Solutions
ATHENA-R1: AI agent for treatment reasoning over a biomedical tool universe
Official Repository of OmniCaptioner
Explore the evolution of AGI through historical context, reasoning models, and agent systems, while gaining hands-on experience with cutting-edge models like Claude 4, DeepSeek-R1, and OpenAI's o3. Learn to critically evaluate AGI benchmarks, understand their limitations, and identify where current models excel or struggle in reasoning tasks.
AI Lawyer is an intelligent reasoning legal assistant powered by DeepSeek , Ollama RAG and LangChain, designed to streamline legal research and document analysis. By leveraging retrieval-augmented generation (RAG), it provides precise legal insights, and contract summarization. With an intuitive Streamlit-based UI, analyze legal documents.
Community registry of LLM serving-path traps that produce confidently wrong measurements: templates, tool parsers, reasoning fields, quant kernel paths, CUDA toolchains, KV allocation, eval harnesses, versioning. Symptom-first, with the check that catches each.
State Sandbox is an experimental game for socioeconomic simulation. It uses Large Language Models (o3-mini) to simulate the world and complex policy impacts.
Pivotal Token Search
To associate your repository with the reasoning-models topic, visit your repo's landing page and select "manage topics."