Testing Local LLM Performance on a MacBook Pro During a Ten-Hour Offline Flight
By
Dmitri Lerko
1mo ago· 4 min readenInsight
85/100
Golden Brown
Bagelometer↗
An everything bagel for the brain. Substantive, layered, well-seasoned.
Score85TypeanalysisSentimentpositive
Summary
An engineer tests the limits of running local LLMs offline on a high-end MacBook Pro (M5 Max, 128GB RAM) during a ten-hour flight from London to Las Vegas. Using Gemma 4 31B and Qwen 4.6 36B models via LM Studio, along with various developer tools and dependencies, they build a billing analytics tool and other engineering projects entirely offline, demonstrating the viability of local AI-assisted development without internet connectivity.
Key quotes
· 4 pulledI flew from London to Google Cloud Next 2026 in Las Vegas. Ten hours with no in-flight wifi.
I used the time to test how far a modern MacBook can carry engineering work on local LLMs alone.
A week old MacBook Pro M5 Max, 128GB unified memory, 40-core GPU.
Gemma 4 31B and Qwen 4.6 36B via LM Studio.
I flew from London to Google Cloud Next 2026 in Las Vegas. Ten hours with no in-flight wifi. I used the time to test how far a modern MacBook can carry engineering work on local LLMs alone.
Setup A week old MacBook Pro M5 Max, 128GB unified memory, 40-cor
