Running Local LLMs With Ollama and Connecting With Python

33 snips

Feb 13, 2026

Christopher Trudeau, Frequent Real Python contributor and PyCoder’s Weekly curator, joins to talk practical Python tooling. They explore running local LLMs with Ollama and connecting them to Python. Short segments cover callable instances with __call__, GeoPandas mapping, subprocess polling changes, peer-to-peer encrypted CLI chat, and a retry library that classifies errors.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

ADVICE

Set Up And Use Tool Calling

Pull the models you need and install the olama Python SDK in a virtual environment before coding.
Define Python functions as tools, pass them to the model, execute selected tools, and append results as tool messages.

ADVICE

Verify Model Sizes And Locations

Check model sizes and available disk space before pulling models (Llama 3.2 ≈2GB, CodeLlama ≈3.8GB).
Look in hidden directories under your home folder for downloaded model files.

INSIGHT

Chat Versus One-Shot Modes

Olama supports both multi-turn chat and one-shot text generation interfaces from Python.
Use messages (role/content dicts) to build conversation context for more accurate outputs.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Would you like to learn how to work with LLMs locally on your own computer? How do you integrate your Python projects with a local model? Christopher Trudeau is back on the show this week with another batch of PyCoder’s Weekly articles and projects.

We cover a recent Real Python step-by-step tutorial on installing local LLMs with Ollama and connecting them to Python. It begins by outlining the advantages this strategy offers, including reducing costs, improving privacy, and enabling offline-capable AI-powered apps. We talk through the steps of setting things up, generating text and code, and calling tools.

We also share other articles and projects from the Python community, including the 2026 Python Developers Survey, creating callable instances with Python’s .__call__(), creating maps and projections with GeoPandas, ending 15 years of subprocess polling, discussing backseat software, a retry library that classifies errors, and a peer-to-peer encrypted CLI chat project.

This episode is sponsored by Honeybadger.

Course Spotlight: Tips for Using the AI Coding Editor Cursor

Learn Cursor fast: Use AI-powered coding with agents, project-aware chat, and inline edits to supercharge your VS Code workflow.

Topics:

00:00:00 – Introduction
00:02:37 – Take the Python Developers Survey 2026
00:03:07 – How to Integrate Local LLMs With Ollama and Python
00:08:15 – Sponsor: Honeybadger
00:09:01 – Create Callable Instances With Python’s .__call__()
00:12:13 – GeoPandas Basics: Maps, Projections, and Spatial Joins
00:16:03 – Ending 15 Years of subprocess Polling
00:18:57 – Video Course Spotlight
00:20:23 – Backseat Software – Mike Swanson
00:39:06 – cmd-chat: Peer-to-Peer Encrypted CLI Chat
00:41:58 – redress: A Retry Library That Classifies Errors
00:43:56 – Thanks and goodbye

News:

Topics:

How to Integrate Local LLMs With Ollama and Python – Learn how to integrate your Python projects with local models (LLMs) using Ollama for enhanced privacy and cost efficiency.
Create Callable Instances With Python’s .__call__() – Learn about Python callables, including what “callable” means, how to use dunder call, and how to build callable objects with step-by-step examples.
GeoPandas Basics: Maps, Projections, and Spatial Joins – Dive into GeoPandas with this tutorial covering data loading, mapping, CRS concepts, projections, and spatial joins for intuitive analysis.
Ending 15 Years of subprocess Polling – Python’s standard library subprocess module relies on busy-loop polling to determine whether a process has completed yet. Modern operating systems have callback mechanisms to do this, and Python 3.15 will now take advantage of these.

Discussion:

Projects:

Additional Links:

Level up your Python skills with our expert-led courses:

Support the podcast & join our community of Pythonistas