← Back to feed

How Much Does It Actually Cost to Run a Local LLM? (Euros per Million Tokens, Measured)

L3 · BuilderTutorials & GuidesTowards Data Science· 7/14/2026

Practical benchmarking data helps builders make informed decisions about local vs cloud inference costs for their AI projects.

AI Summary

A detailed cost analysis comparing the actual energy consumption and operational costs of running various local LLMs on an RTX 3090 GPU versus cloud API pricing. The author measured real electricity costs through precise power sampling and found surprising results - smaller models aren't always cheaper, and three local models actually cost more than cloud alternatives. The article provides concrete measurements in euros per million tokens across eight different models.

Excerpt

I measured the actual GPU electricity for eight local models on one RTX 3090 — and the cheapest wasn't the smallest, nor the priciest the biggest. The post How Much Does It Actually Cost to Run a Local LLM? (Euros per Million Tokens, Measured) appeared first on Towards Data Science.

Read Original
0 upvotes · 0 downvotes · 1 min read

Related Articles

L3 · BuilderTutorials & GuidesTowards Data Science
Introduction to Reinforcement Learning: Multi-Armed Bandit Simulation in Python

This tutorial teaches reinforcement learning fundamentals through building a multi-armed bandit simulation in Python, using a classic RL problem to demonstrate trial-and-error learning.

L4 · DeveloperTutorials & GuidesTowards Data Science
How to Make a Cloud Read a Drone's Mind (and Cut Data Usage by 94%)

A tutorial on building world models in PyTorch to reduce drone/robot data transmission by 94% using predictive modeling instead of constant sensor updates.

L3 · BuilderTutorials & GuidesSimon Willison's Blog
Using Parseable with Datasette for OpenTelemetry traces

Simon Willison shares how to use Parseable observability tool with Datasette for OpenTelemetry traces, including setup patterns and integration examples.

L2 · PractitionerTutorials & Guides@bcherny
Boris Cherny (@bcherny): I am surprised that people are surprised this is how I prompt Claude. Talk to Claude the way you would a coworker. There's no secret to prompting. There's no need to be overly scaffolded or prescript

Boris Cherny advocates for natural conversation with Claude rather than complex prompting techniques, suggesting users treat it like a coworker.

L2 · PractitionerTutorials & GuidesLessWrong AI
Comparing AI Policy Courses

Compares AI policy courses from BlueDot Impact, Lens Academy, Vista Institute, and Center for AI and Digital Policy for career upskilling.