Skip to content
View original post on X: Yuchen Jin
Published
44/100

DeepSeek-V4.1-Flash-1017 scores 77 on DeepSWE, claims Opus-level results

AIYuchen Jin says DeepSeek-V4.1-Flash-1017 scores 77 on DeepSWE v1.1, matching Opus 5-level performance at 83x lower cost. He adds that Databricks will launch the model.
Yuchen JinVerified on X
@Yuchenj_UW
View on X

The "OSS will never catch up" takes didn't survive a week.

DeepSeek-V4.1-Flash-1017 scores 77 on DeepSWE v1.1. Opus 5-level performance at 83x lower cost. Incredible for a model of this size. We will be launching it on Databricks!

Never bet against open source.

Source details

Yuchen Jin · x.comPublished

Original title: The "OSS will never catch up" takes didn't survive a week.

Topics

DeepSeekOpen-source ecosystemChina AIModel releasesEvals & benchmarks