Model Releases

Fireworks benchmarks DeepSeek-V4.1-Flash: DeepSWE matches GPT-6 Astra at 1/15 the cost per task

Original title:DeepSeek-V4.1-Flash on Fireworks: Astra-level DeepSWE at 1/15th the cost

AI summary

Fireworks ran a full benchmark suite on DeepSeek-V4.1-Flash, released last week, finding it reaches GPT-6 Astra-level coding accuracy on DeepSWE with 74.34% pass@1 at $0.43 per task, about 1/15 of Astra's cost.

Key Takeaways

  • Fireworks ran a full benchmark suite on DeepSeek-V4.
  • 1-Flash,released last week,finding it reaches GPT-6 Astra-level coding accuracy on DeepSWE with 74.
  • 34% pass@1 at $0.
  • 43 per task,about 1/15 of Astra's cost.

Quick Facts

Published
Updated
Source
Fireworks AI(网页)
Category
Model Releases
AI score
72
Read the original Read the full article at the original source
Why it matters

原文给出 DeepSeek-V4.1-Flash 与多款前沿模型的实测成本与准确率数据,可帮助读者按任务匹配合适模型。

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: Fireworks benchmarks DeepSeek-V4.1-Flash: DeepSWE matches GPT-6 Astra at 1/15 the cost per task. September 14, 2026. https://new.nat6.net/en/news/ltzy7o99txg9o69cu70prdnrr/

View all →