---
format: "aidr-story-markdown/v1"
id: "da9c62c52f6c35e6916b07245e2a0775f3119e140ea4c4101fd2bd36f52592d3"
canonical_url: "https://aidr.today/da9c62c5?lang=en"
title: "Quoting Ted Sanders"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-28T23:23:21.000Z"
category: "Models"
topics: ["openai","llm"]
source_urls: ["https://simonwillison.net/2026/Oct/28/ted-sanders/"]
summary: "Ted Sanders explained that model size is the primary factor driving LLM performance variation, with larger models generally being more performant, costly, and knowledgeable, and that version numbers reflect release order, relationships to other releases, and improvement magnitude."
---

# Quoting Ted Sanders

> [Open the canonical story](<https://aidr.today/da9c62c5?lang=en>)

**Published:** 2026-10-28T23:23:21.000Z
**Category:** Models
**Topics:** openai, llm

## Summary

Ted Sanders explained that model size is the primary factor driving LLM performance variation, with larger models generally being more performant, costly, and knowledgeable, and that version numbers reflect release order, relationships to other releases, and improvement magnitude\.

## Sources

- [Story source](<https://simonwillison.net/2026/Oct/28/ted-sanders/>)