terra beats Sonnet on a report run: 36s and $1.22 vs 81s and $4.03
samuelcolvin · x · 2026-07-24
- terra finished the report in 36 seconds, versus 81 seconds for Sonnet.
- The reply links the gap to Sonnet “digging for too long,” suggesting extra reasoning/search steps may be driving both latency and cost.
- Report cost was also much lower on terra: $1.22 vs $4.03 on Sonnet.
Related event: Terra Outperforms Sonnet in Lead Classification at Lower Cost(3 posts)→
More from Models
- Anthropic's Terra returns with better writing and 2X usage vs 6.1 Sol — weswinder · 2026-10-27
- User catches model failing to parse 400-paper file, citing stale news sources instead — jungofthewon · 2026-10-27
- Odd eval result: Gemini 4.1 Flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27