SWE-1.7

Released 8 Jul 2026 · cognition.com

Benchmark results

Benchmark	Score
Terminal-Bench 4.0best score across reasoning efforts · reported later, in Introducing SWE-2: Pushing the Pareto Frontier	7.6%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	42.3%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.
best score across reasoning efforts · reported later, in Introducing SWE-2: Pushing the Pareto Frontier	42.0%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.
DeepSWE v1.1best score across reasoning efforts · reported later, in Introducing SWE-2: Pushing the Pareto Frontier	37.7%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.
Terminal-Bench 2.1max reasoning effort; Devin CLI harness; Cognition's own eval framework; 4h timeout	81.5%Corroborated The same number is printed in 1 other lab document: 1

Sources

Reported by Cognition