SWE-1.7

Name: SWE-1.7
Author: Cognition

CognitionProprietaryPending Human Review

SWE-1.7 is Cognition’s Kimi K2.7 Code-derived model for long-running software-engineering tasks, released through Devin Web, Desktop, and CLI. Extensive reinforcement learning targets thorough codebase exploration, faithful task execution, and concise reasoning while preserving the ability to think longer on hard problems. Learned self-compaction summarizes working state and resumes from those summaries, allowing agent sessions to continue beyond one model context window without implying an unlimited single-call context. Training uses hardened executable environments and alternating cost penalties. The publisher explicitly distinguishes its behavior from the base checkpoint. Cognition does not publish weights or verified standalone model limits for this release.

2026-07-08

Kimi-derived Mixture-of-Experts Transformer

Proprietary

Compare with other models

Specifications

Architecture: Kimi-derived Mixture-of-Experts Transformer
License: Proprietary
Type: text
Modalities: text

Benchmark Scores

Advanced Specifications

Model Family: SWE
Finetuned From: Kimi K2.7 Code
API Access: Not Available
Chat Interface: Available
Variants: SWE-1.7 Lightning

Capabilities & Limitations

Capabilities: long-horizon codingself-compactiontool usecodebase exploration
Known Limitations: Integrated product access does not establish a general model inference APIEvaluation results depend on harnesses, task sets, and effortGenerated changes require execution and reviewSelf-compaction can discard useful contextTraining may encourage changes beyond the narrowest task scope
Notable Use Cases: asynchronous engineering tasksrepository bug fixesmulti-file implementation
Function Calling Support: Yes
Tool Use Support: Yes

Resources

Related Models

SWE-2

Cognition

SWE-2 is Cognition’s Kimi K3-derived coding model, first publicly introduced and served in Devin Desktop and CLI on September 10, 2026. It extends Cognition’s reinforcement-learning recipe to the roughly 2.8T-parameter scale, optimizing several reasoning-effort levels in one training run. Effort-specific cost penalties target the tradeoff between engineering capability and inference cost, while expanded coding environments and hardened verifiers encourage more reliable task completion. The released identity is the trained model rather than the surrounding Devin application. Web and Fusion rollout were announced separately. Public weights and exact model-serving context/output limits are not established by the launch announcement.

Typetext

ParametersApproximately 2.8T total

2026-09-10

Proprietary

Details Compare

SWE-check

Cognition

SWE-check is a specialized bug-detection language model developed by Cognition with Applied Compute and first made available in the Windsurf Next public preview. It analyzes a change diff, investigates the surrounding code, and produces structured bug descriptions and fixes for display in the editor. Reinforcement learning first builds task capability, then tunes latency and product behavior using a separate alignment phase. This is a distinct trained generative review model rather than a generic code-review feature or binary decision endpoint. Architecture, parameter counts, and model context limits remain undisclosed. Its reported parity with frontier models applies to internal in-distribution evaluations, with a remaining gap on out-of-distribution tasks.

SWE-1.6

Cognition

SWE-1.6 is Cognition’s next coding model, first served as a limited-user preview on March 1 and generally released in Windsurf on April 7, 2026. It uses the same underlying pretrained model as SWE-1.5, but the production release was post-trained anew to improve both coding capability and interaction behavior. Training discourages overthinking, repeated reasoning loops, and unnecessary terminal use, while encouraging parallel tool calls. The earlier preview and production checkpoint are grouped under this identity, with their behavioral differences made explicit. Standard and faster serving options provide the same model intelligence; neither establishes publicly downloadable weights or a standalone model API.