Model Releases

DeepSeek upgrades deepseek-chat to DeepSeek-V2.5-1210

Original title:deepseek-chat

AI summary

DeepSeek upgraded the deepseek-chat model to DeepSeek-V2.5-1210, improving capabilities across the board. Math capability on the MATH-500 benchmark rose from 74.8% to 82.8%, and code capability on the LiveCodebench (08.01 - 12.01) benchmark rose from 29.2% to 34.38%. The new version also improves the user experience of file upload and web page summarization features.

Key Takeaways

  • DeepSeek upgraded the deepseek-chat model to DeepSeek-V2.
  • 5-1210,improving capabilities across the board.
  • Math capability on the MATH-500 benchmark rose from 74.
  • 8%,and code capability on the LiveCodebench (08.
  • 01) benchmark rose from 29.

Quick Facts

Published
Updated
Source
DeepSeek:API 更新日志
Category
Model Releases
AI score
67
Read the original Read the full article at the original source
Why it matters

DeepSeek把deepseek-chat数学拉到82.8%,代码涨了5个点,做复杂推理的开发可以试试水,但这还是V2.5系列的常规升级,别当次代际跃迁。

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: DeepSeek upgrades deepseek-chat to DeepSeek-V2.5-1210. December 10, 2024. https://new.nat6.net/en/news/cms8ndnte01pwrof179eenw2m/

View all →