DeepSeek V4.1 Flash Matches Opus 5 Benchmarks at 1/83 Cost

Original title: The "OSS will never catch up" takes didn't survive a week.
68AI Score

Open-source model DeepSeek-V4.1-Flash-1017 scored 77 on the DeepSWE v1.1 benchmark, approaching Opus 5-level performance at 1/83rd the cost. This result challenges the notion that open source cannot catch up and is slated for launch on Databricks.

Full Report

Yuchen Jin (@Yuchenj_UW) · 2026-10-17 17

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.