reasoning-agent
TxAgent
TxAgent: An AI Agent for Therapeutic Reasoning Across a Universe of Tools
- 656
- Python
- MIT
- 16 mo ago
KnowAgent
[NAACL 2025] KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents
- 266
- Python
- Apache-2.0
- 22 mo ago
ZoomEye
[EMNLP-2025 Oral] ZoomEye: Enhancing Multimodal LLMs with Human-Like Zooming Capabilities through Tree-Based Image Exploration
- 96
- Python
- Apache-2.0
- 3 wk ago
thinking-intervention
Used for thinking process intervention of reasoning models such as DeepSeek-R1, effectively controlling the reasoning thinking process. 用于DeepSeek-R1等推理模型的思维过程干预,有效控制推理思考过程
- 24
- Python
- Apache-2.0
- 20 mo ago
sheldon
SHELDON is a reasoning harness that searches observations and hypotheses as two separate spaces, focusing on finding unknown observations to generate new hypotheses.
- 8
- Python
- MIT
- 5 wk ago
video-analysis-agent
This project implements an AI agent that verifies if automated Hercules test runs were executed as intended by comparing planning logs, video evidence, and final outputs. It uses open-source LLMs and computer vision tools to flag deviations, providing detailed reports with technical insights.
- 4
- Python
- 17 mo ago
LangGraph-Framework-Basic-to-Advanced
LangGraph is a powerful framework built on LangChain that enables the creation of stateful, multi-step, and agentic workflows using directed graphs. It simplifies complex LLM orchestration by allowing conditional branching, memory, and tool integrations in a visual and modular way.
- 2
- Jupyter Notebook
- MIT
- 7 d ago
Reasoning-Enabled-AI-assistant-using-DeepSeek-and-Gemini-Flash
AI Powered Python Coding Assistant with Reasoning/Thinking feature is a Streamlit-based web application that leverages Google Gemini and Groq (DeepSeek-LLaMA) models to assist developers with Python code generation, debugging, and optimization.
- 2
- Python
- MIT
- 19 mo ago
RL-Symbolic-Feedback
Enhancing reasoning capabilities of LLMs by fine-tuning based algorithm using Symbolic AI feedback component
- 2
- Jupyter Notebook
- 2 y ago
llama-3.2-3b-reflect
Llama-3.2-3B-Instruct distilled on DeepSeek-R1 reasoning traces to think out loud. GSM8K accuracy falls from 75.0% to 65.0% while output length more than doubles — published as a measured negative result, with the per-problem breakdown behind the headline number.
- 1
- Jinja
- MIT
- 4 wk ago
RayCodes_Granite_4.2
IBM Granite 4.2: The First 2GB Local AI With Adjustable Thinking! Enterprise open-source dense reasoning models with dynamic CoT toggle and native Ollama tool calling.
- 1
- Python
- 3 mo ago









