#reasoning effort levels

Cognition Releases SWE-2: A Kimi K3 Post-Trained Coding Model That Matches Fable 5.1 on FrontierCode at 64% Lower CostNEW⭐7

Cognition's SWE-2 coding model, post-trained from Kimi K3, hits 50.0% on FrontierCode 1.1, matching Fable 5.1 at 64% lower cost and running only inside Devin.