SWE-2 (Reasoning)
by Cognition
SWE-2 (Reasoning) is a reasoning model from Cognition, released on September 10, 2026. It ranks #31 of 247 models by Intelligence Score with a score of 47.9.
SWE-2 is Cognition's most advanced coding model, post-trained from Moonshot AI's 2.8-trillion-parameter Kimi K3 with reinforcement learning scaled to the multi-trillion-parameter regime for the first time. A single RL run trains all reasoning-effort levels (medium, high, max) together with per-effort cost penalties tuned to the base model's Pareto frontier, advancing the whole cost-performance curve rather than a single operating point. On FrontierCode 1.1 Main it scores 50.0%, within one point of Claude Fable 5.1 (50.9%) while Cognition states it is 64% cheaper; on DeepSWE 1.1 it scores 73.0%; on Terminal-Bench 2.1 it scores 92.8%; on Terminal-Bench 4.0 it scores 27.3%. SWE-2 is available starting 2026-09-10 in Devin Desktop and Devin CLI, with Devin Web and Fusion rollout in progress; Cognition has not published a standalone public API, model card, or context-window specification independent of the Devin product.
Key specs
| Provider | Cognition |
|---|---|
| Release date | 2026-09-10 |
| License | proprietary |
| Intelligence Score | 47.9 |
| Coding index | 38.7 |
Quick answers
Is SWE-2 (Reasoning) open source?
No. SWE-2 (Reasoning) is a proprietary model from Cognition (proprietary license).
When was SWE-2 (Reasoning) released?
Cognition released SWE-2 (Reasoning) on September 10, 2026.
How does SWE-2 (Reasoning) compare to Claude Opus 5.5 (Reasoning)?
SWE-2 (Reasoning) scores 47.9 on the Intelligence Score versus 59.9 for Claude Opus 5.5 (Reasoning), the current #1.