SWE-1.7: Frontier Intelligence at a Fraction of the Cost | Cognition
Today, we’re launching SWE-1.7, the most capable model we’ve trained so far. It reaches frontier-level intelligence at a much lower cost, advancing the cost-performance Pareto curve.
Ben Pan*, Carlo Baronio*, Rohan Choudhury, Eric Lu, Ryan Kim, Deniz Birlikci, TC Qin, Sam Lee, Fermi Ma, Allen Liu, Yang Liu, Sampriti Panda, Jacob Teo, Ray Wang, Gary Chang, Steven Cao, Silas Alberti*equal contribution 07.08.26 Today, we’re launching SWE-1.7, the most capable model we’ve trained so far. It reaches frontier-level intelligence at a much lower cost, advancing the cost-performance Pareto curve. SWE-1.7 is the result of broad improvements across our RL pipeline: better infrastructure, more stable training, higher-quality data, and new techniques for long-horizon tasks. Since…
saved by
related reading
- Introducing SWE-2: Pushing the Pareto Frontiercognition.com
- Introducing SWE-grep and SWE-grep-mini: RL for Multi-Turn, Fast Context Retrieval | Cognitioncognition.ai
- Scaling Agentic RL: 365,000+ Environments for SWE, Terminal, and Searchprimeintellect.ai
- Kevin-32B: Multi-Turn RL for Writing CUDA Kernels | Cognitioncognition.ai
- FrontierCode 1.1 | Cognitioncognition.com
- FrontierSWEfrontierswe.com
- Is Frontier Asynchronous RL Solved? — Luke J. Huangluk-huang.github.io
- frontier model training methodologies | Alex Wa's Blogdjdumpling.github.io
- FrontierSWEfrontierswe.com
- [2607.00053] SWE-Router: Routing in Multi-turn Agentic Software Engineering Tasksarxiv.org
- Composer2.pdfcursor.com
- 2502.18449arxiv.org