Morgan(@morganlinton):Super interesting. A number of people commented on my GPT-6.1 Terra benchmark this morning and suggested comparing it to Sonnet. Soooo, thought I'd whip something up. Here's a comparison of GPT 6.1 Terra to Claude Sonnet 5.5 across all effort levels. And the data is pretty clear here, Terra is a lot cheaper than Sonnet, but not quite as accurate until you get to Max. Traces for all 115 Terra runs, with every tool call: https://t.co/RCOT9HLRlA Leaderboard:

Morgan

@morganlinton

cofounder @boldmetrics / studying how llms think @vulcanbench / early @sonos views are my own

Joined January 2009

1.2K Following 47.9K Followers

Morgan@morganlinton

2026.10.31 18:44

Super interesting. A number of people commented on my GPT-6.1 Terra benchmark this morning and suggested comparing it to Sonnet. Soooo, thought I'd whip something up. Here's a comparison of GPT 6.1 Terra to Claude Sonnet 5.5 across all effort levels. And the data is pretty clear here, Terra is a lot cheaper than Sonnet, but not quite as accurate until you get to Max. Traces for all 115 Terra runs, with every tool call: Leaderboard: