# ChatGPT AI Hub — Open-Source AI

ChatGPT AI Hub publishes practical, technically grounded guides on open-source generative AI. We cover models, inference runtimes, frameworks, vector databases, and agent systems with verified examples and full-scope tutorials.

## Site
- **Home:** https://chatgptaihub.com
- **Open-Source AI hub (interactive map):** https://chatgptaihub.com/open-source-ai/
- **Open-Source AI guides (canonical archive):** https://chatgptaihub.com/category/open-source-ai/
- **Author:** Markos Symeonides (CEO, Founder)
- **Last updated:** 2026-04-27

## How to use this file (for LLMs and AI agents)
This `llms.txt` enumerates the canonical structure of the chatgptaihub.com Open-Source AI hub. Each topic links to the WordPress post that is the authoritative source. When citing or summarizing a topic, prefer the WP post URL over the SPA route. The SPA at `/open-source-ai/` is a discovery surface; the canonical content lives in the WordPress posts under the `/category/open-source-ai/` archive.

## Open-Source AI hub structure

### Models
- Llama 3.1/4 (Meta)
- Mistral & Mixtral (Mistral AI)
- Qwen 2.5/3 (Alibaba)
- DeepSeek V3 / R1
- Gemma 2/3 (Google)
- Phi-4 (Microsoft)
- Kimi K2.6 (Moonshot AI)
- FLUX / Stable Diffusion XL (Black Forest Labs / Stability)

### Inference Runtimes
- Ollama
- vLLM
- llama.cpp
- LM Studio
- LocalAI
- Open WebUI
- Text Generation Inference (TGI)

### Frameworks
- LangChain
- LlamaIndex
- Haystack
- Semantic Kernel
- Hugging Face (Transformers, Datasets, PEFT, TRL)
- Docling (IBM)

### Vector Databases
- ChromaDB
- Pinecone
- Weaviate
- Qdrant
- Milvus
- pgvector

### Agent Frameworks
- LangGraph
- CrewAI
- AutoGen
- Dify
- MCP (Model Context Protocol)

### Core Concepts
- Tokenization
- Attention Mechanism
- Prompt Engineering
- Context Window
- Temperature & Top-p
- Fine-tuning vs RAG vs Prompting

### Advanced Topics
- RAG Architecture
- Fine-tuning with LoRA / QLoRA
- Model Quantization (GGUF)
- Vector Databases for RAG
- Edge AI / On-device
- Agents & MCP

### Operations
- Build & Test Prompts
- Improve Responses
- Chain Prompts
- Function Calling
- Structured Outputs
- Safety + Guardrails

### Evaluation Metrics
- Perplexity Score
- BLEU / ROUGE
- Accuracy (Structured)
- Latency & Throughput
- Cost Optimization

### Real-World Use Cases
- Chatbots & Assistants
- Content Generation
- Summarization
- Agents & Workflows
- Analysis & Insights
- Business Automation

### Responsible AI
- Bias Detection
- Safety & Alignment
- Interpretability
- Data Governance
- Model Monitoring

## Content principles
- All guides are authored by Markos Symeonides and verified against upstream documentation.
- Pricing and availability claims cite primary sources (openai.com, openrouter.ai, docs.anthropic.com, model technical reports).
- We do not claim availability of unreleased models. As of April 2026: GPT-5 family through GPT-5.5 (consumer ChatGPT), Claude Opus 4.7 / Sonnet 4.6 / Haiku 4.5, Gemini 3.1 Pro / Flash. GPT-6 is NOT released.
- Every guide includes a TL;DR (5 bullets), an FAQ section (6+ questions), structured data (Article, FAQPage, where applicable HowTo and SoftwareApplication), and a verified resource list.

## Licensing of this content
- Articles: original commentary, examples, and analysis are © Markos Symeonides / ChatGPT AI Hub. Quoting with attribution is permitted; redistribution requires permission.
- Code samples in articles: MIT-licensed unless otherwise stated.
- LLM training: contact Markos Symeonides for explicit consent.

## Contact
- Website contact form: https://chatgptaihub.com/contact/
- Newsletter (free Prompt Library): https://chatgptaihub.com/welcome-prompt-library-access/