Day 7 receipt
GPT-6.1 Terra
Day 2 of 7OpenAI's fast, lower-priced GPT-6.1 model for computer use, visual reasoning and everyday work tasks, available in ChatGPT Work, Codex and the API.
Launch dayWhat they claimed, quoted word for word · 4
“At max effort, Terra is over three times as token efficient as Claude Opus 5.5, and one-fourth its price per token, resulting in a ~9x lower cost per task.”
No independent cost-per-task comparison against Opus 5.5 has appeared yet. One early user run found max effort expensive relative to Sol.
“Terra is also much quicker, completing tasks over three times faster than both GPT-6 Astra and GPT-6.1 Sol.”
No first-hand speed measurements yet.
“At maximum reasoning effort, GPT‑6.1 Terra solves 50.3% of tasks, more than double GPT‑5.6 Terra’s published score of 21.5%.”
A vendor benchmark number; no independent rerun yet.
“Standard API pricing is $1 per million input tokens, $0.05 per million cached input tokens, and $5 per million output tokens.”
List prices, not tested. One user's benchmark site suggests max effort costs more per task than Sol, while medium costs less.
Day 7What real users report · 1
A user running their own benchmark said Terra at max effort scored below Sol and cost about twice as much, while medium effort looked like the best use and cheaper than Sol.
Comment on a 627-point launch thread with 420 comments · Open the source