Frontier Log

Cognition

SWE-1.6

Released · cognition.com

Price

Per 1M tokensInputCachedOutput
Self-serve extra-usage rate in Devin Desktop and Devin CLI$0.50$0.20$2.50
Price history
DateInputCachedOutput
As of docs.windsurf.com · 2 · Corroborated$0.50$0.20$2.50

Benchmark results

BenchmarkScore
SWE-bench ProPublic set (731 tasks) · 731 tasks; harness not stated50.4%Checked A person checked this number against its source on 7 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.
FrontierCode 1.1Main set · max reasoning effort · reported later, in SWE-1.7: Frontier Intelligence at a Fraction of the Cost9.4%Checked A person checked this number against its source on 7 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.
Terminal-Bench 2.1max reasoning effort; Devin CLI harness; Cognition's own eval framework; 4h timeout · reported later, in SWE-1.7: Frontier Intelligence at a Fraction of the Cost39.7%Checked A person checked this number against its source on 7 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.
Sources

Reported by Cognition

Related

All models