Topic
Open-Source & Local Models in Frameworks (Llama 4, Mistral, DeepSeek, Qwen)
14 posts on open-source & local models in frameworks (llama 4, mistral, deepseek, qwen) — part of framework tutorials on the n4n AI blog.
Serving Llama 4 with vLLM and an OpenAI-compatible API
A practical guide to serving Llama 4 with vLLM behind an OpenAI-compatible API, covering hardware requirements, quantization choices, deployment patterns, and common production pitfalls.
Route local models with LiteLLM: Qwen3 and Llama 4
Configure LiteLLM to route requests between local Qwen3 and Llama 4 models with fallback, load balancing, and cost tracking.
Qwen2.5-72B in LangChain: a complete setup guide
Complete guide to integrating Qwen2.5-72B with LangChain — local Ollama setup, OpenAI-compatible endpoints, streaming patterns, and production pitfalls.
Mistral Large in Semantic Kernel: a .NET integration guide
A practical guide to integrating Mistral Large with Semantic Kernel in .NET, covering setup, chat completion, streaming, function calling, and common pitfalls.
LlamaIndex plus Ollama: indexing docs with DeepSeek-V3
Build a local-first document indexing pipeline with LlamaIndex, Ollama embeddings, and DeepSeek-V3 via OpenAI-compatible API.
Haystack pipelines with self-hosted DeepSeek-V3
Build a production-ready Haystack RAG pipeline with self-hosted DeepSeek-V3 using vLLM and Ollama, with working code and deployment patterns.
CrewAI agents powered by local Qwen2.5 models
Run CrewAI multi-agent workflows locally with Qwen2.5 models using Ollama — complete setup, configuration, and verification steps.
Integrate Mistral Small 3.1 into a LlamaIndex RAG pipeline
Build a production-ready RAG pipeline with Mistral Small 3.1 and LlamaIndex, including document ingestion, retrieval, and generation with runnable code.
Fine-tune Llama 4 Scout with Hugging Face Transformers
A step-by-step guide to fine-tuning Llama 4 Scout with Hugging Face Transformers, covering LoRA, quantization, and distributed training.
DeepSeek-R1 reasoning chains in LangGraph: a walkthrough
Build a DeepSeek-R1 reasoning chain in LangGraph with streaming, state management, and fallback handling — complete runnable code included.
Run Llama 4 Maverick locally with Ollama and LangChain
Practical steps to run Llama 4 Maverick with Ollama and LangChain locally, from Ollama install to verified streaming chat in Python.
Ollama vs vLLM for serving Llama 4 in production apps
Practical comparison of Ollama vs vLLM for serving Llama 4 in production: capabilities, cost, latency, ergonomics, limits, and which to choose.
Deploy DeepSeek-V3 with vLLM behind a LangChain agent
Step-by-step tutorial: deploy DeepSeek-V3 with vLLM and connect it to a LangChain agent on self-hosted local hardware via OpenAI-compatible API.
Building a local RAG app with Mistral 7B and Ollama
Build a fully local RAG application using Mistral 7B and Ollama: install models, ingest documents, embed with nomic, retrieve, and generate offline.
More topics in framework tutorials
- Building Chatbots with Vercel AI SDK & Next.js15
- LangChain RAG with Vector Databases15
- LangGraph Multi-Agent Workflows15
- LlamaIndex Query Engines for RAG15
- AutoGen Multi-Agent Conversations & Group Chat14
- Choosing an AI Framework by Use Case14
- CrewAI Agent Roles & Task Design14
- CrewAI Real-World Crew Examples14
- Haystack RAG Pipelines14
- LangChain Agents & Tool Calling14
- LangChain Multi-Model Fallback & Routing14
- LangChain.js for Node & TypeScript14