The home for great developer writing.

We surface the best developer writing from thousands of independent blogs, updated daily.

I'm (mostly) picking models on speed now, not intelligence

· Martin Alderson · Aug. 2, 2026, 1:04 a.m.

Machine Learning productivity AI Models Performance Metrics

Summary

Martin Alderson discusses his shift in evaluating AI models based on speed (tokens per second) rather than raw intelligence, suggesting that higher speed may become more critical than intelligence in the future of AI model selection.

Read full post on martinalderson.com →

MORE POSTS LIKE THIS

Try the very fast models

Natemeyvis · Jul 27, 2026

future of work Generative AI

We Added Jev as a Reranker. Here's What We Learned.

Hindsight Blog · Sep 24, 2026

hindsight Agent Memory

Claude Opus 5.5: The System Card

Thezvi Wordpress · Sep 23, 2026

AI artificial-intelligence

Your Agents Should Have Names

Daniel Miessler · Sep 21, 2026

Machine Learning productivity

Feature Request: Dedicated Premium Requests for GPT-6 Astra with Auto-Fallback to GPT-5.6 Sol on 0%

Community Openai · Sep 20, 2026

productivity developer experience

GPT and Claude go to heraldry school

Thoughtbot · Aug 26, 2026

Machine Learning AI Models

Discover more posts →

AUTHOR

Advertise

Sponsor diff.blog

Put your product in front of developers who read and write about their craft. One exclusive sponsor at a time.

Become a sponsor →