From the source
Lead story
Ask your AI
Top stories
Models & availability
Latest
Introducing SWE-2: Pushing the Pareto Frontier
SWE-2 achieves 50% on FrontierCode 1.1, 64% cheaper than Fable 5.1
Cognition introduced SWE-2, a coding model that achieves 50.0% on FrontierCode 1.1 Main, within one point of Fable 5.1 while being 64% cheaper.
SWE-2 is post-trained from Kimi K3, a 2.8T-parameter model, and uses a cost-penalized RL algorithm to train all reasoning-effort levels in a single run.
The model is available starting today in Devin Desktop and CLI, and is rolling out on Devin Web and Fusion.
From the source
Today we’re introducing SWE-2, our most advanced coding model yet.
cognition.com
Other stories
Amazon · 2 OctSweep thousands of leases for compliance using Amazon Quick and the Adjudicated Query patternAmazon · 2 OctAdd secure Web Search to Claude Desktop with Amazon Bedrock AgentCoreAmazon · 2 OctFine-tune a search agent with multi-turn RL on Amazon SageMaker AI