A collection of guides and examples for building intelligent applications with open models on Nebius Token Factory.
⭐ If you find this repo useful, give it a star! You'll be notified of new updates and help others discover it too — thank you!
- Token Factory Workshop
- Office Hours • Events supported by Nebius AI
- LangChain deep agent examples
- Community contributions
- Post training examples
- Distillation guide
- Cool Apps / Demos
Network, learn, and build with fellow builders - including hackathons, workshops, and community events.
Join Office Hour online.
Events / hackathons / workshops
- Prerequisites
- A Nebius Token Factory account and API key — sign up for free here
- Python runtime (local, Google Colab, etc.)
- Follow the getting started guide
- Follow project-specific instructions in each example's README
🎁 Models
Checkout the latest model guides and sample code.
Featuring: Kimi-K3 • DeepSeek V4 Pro • GLM-5.2 • Qwen3.5-397B-A17B • Nemotron family
📘 APIs
Access Token Factory using various APIs:
OpenAI-compatible API • LiteLLM • ai-suite • Llama-index
🎠 Agents
Numerous agent examples ranging from starter to intermediate and advanced.
Featured AI Agent frameworks:
CrewAI
• 
Agno
• 
LangChain
• 
Google ADK
• 
Llama-index
• 
Pydantic
• 
AWS Strands
🔍 RAG
| Example | Description | Tech Stack |
|---|---|---|
| PDF RAG | Simple PDF RAG application | LlamaIndex + Nebius AI |
| Chat with Documents | Web app to chat interactively with PDF documents | LlamaIndex + Nebius AI + Streamlit |
| Internal Content Generation Platform | Create social posts, articles, and demos with a RAG pipeline | Qdrant + Nebius AI + Qwen3-Embedding |
| End-to-end RAG Pipeline for PDFs | Process and query PDF documents using vector DB and open-source embeddings | LlamaIndex + Milvus + Nebius AI + Qwen3-Embedding + GPT-OSS |
| Internal Support Agent | Multi-agent support agent with web search, document RAG, Notion/Calendly tools, and Slack bot | Nebius AI + LangGraph + Weaviate |
⛓️ Integrations
Guides for connecting third-party tools to Token Factory:
- OpenClaw — self-hosted AI agents on open models
- Tavily — LLM-optimized search API for agentic research
- Pixeltable — multimodal AI data infrastructure with native Token Factory chat + embeddings
See also the official catalog: docs.tokenfactory.nebius.com/integrations
| Example | Description | Tech Stack |
|---|---|---|
| Simple function calling | Demonstrates how to call functions | Nebius AI |
Post-training lets you customize and improve pre-trained language models for your specific use cases through fine-tuning and other optimization methods.
| Name | Description | Tech Stack |
|---|---|---|
| Distillation 1 | Example of a distilled model for grammar checking | Nebius AI |
Hands-on, guided learning sessions:
Have some fun with models:
We welcome your contributions! Open issues, submit pull requests, and share your experience.
🧑🏻🤝🧑🏼 View community contributions
This project is licensed under the MIT License. See the LICENSE file for details.
Thanks to all of our amazing contributors!
© Nebius BV, 2025