Introducing SWE-2, our closest model yet to the frontier. On leading evals, it scores on par with recent frontier models – at up to 70% lower cost. We scaled RL to multiple trillions of parameters, with a refined recipe that pushes the Pareto curve on both capabilities & cost. t.co/FjBxrikExs

- Views
- 2,303,367
- Likes
- 6,778
- Replies
- 440
- Reposts
- 511
- Quotes
- 523
- Bookmarks
- 1,750
Introducing SWE-2, our closest model yet to the frontier.
The post announces SWE-2 as a high-performing model that approaches frontier results at a lower cost. Its comparison table shows SWE-2 beside other models across three evaluation benchmarks, with the featured column highlighted and its strongest result emphasized.
- Views
- 2.3M
- Likes
- 6.8K
- Replies
- 440
- Reposts
- 511
- Quotes
- 523
- Bookmarks
- 1.8K
- Topics
Analysis
The post announces SWE-2 as a high-performing model that approaches frontier results at a lower cost. Its comparison table shows SWE-2 beside other models across three evaluation benchmarks, with the featured column highlighted and its strongest result emphasized.
Sign in to see the full analysis.
Sign inIntroducing SWE-2, our closest model yet to the frontier. On leading evals, it scores on par with recent frontier models – at up to 70% lower cost. We scaled RL to multiple trillions of parameters, with a refined recipe that pushes the Pareto curve on both capabilities & cost. https://t.co/FjBxrikExs
Introducing SWE-2, our closest model yet to the frontier.
- 01The Announcement
Introduce a new offering and make a strong claim about how close it is to the leading standard.
- 02The Proof Point
Support the claim with benchmark results and a clear cost advantage.
- 03The Explanation
Explain the underlying scale or process improvements that produced the performance and cost gains.
Reusable caption template.
Sign in to see the full analysis.
Sign in