DeepSeek V3-0324 Becomes Top Non-Reasoning AI Model

Posted by By Dan Taylor 2 Min Read

China’s DeepSeek has released a new open weights AI model that just topped the charts in a major performance category.

The new model, DeepSeek V3-0324, is now the highest-scoring non-reasoning model on the Artificial Analysis Index, beating out proprietary models like Gemini 2.0 Pro, Claude 3.7 Sonnet, and Grok 3.

This marks a first for open models — DeepSeek V3-0324 is the top performer in its class, ahead of well-funded closed-source competitors.

Strong Performance With Open Weights

V3-0324 scored 53 on the Artificial Analysis Index, a 7-point jump from its previous version. That increase shows DeepSeek is not just catching up to the field, but moving ahead.

The model holds its own despite no major hardware breakthroughs or dramatic changes in architecture. It uses:

  • 128k context window (limited to 64k on DeepSeek’s own API)
  • 671 billion total parameters
  • FP8 precision
  • MIT license
  • Text-only output

This level of performance from an openly available model is drawing attention across the AI industry.

Not Even the Flagship Model

This release is more surprising because DeepSeek V3-0324 isn’t even the company’s top model.

According to DeepSeek, model R1 still performs better overall. A new version, R2, is reportedly on the horizon.

Anyone can try DeepSeek V3-0324 now at chat.deepseek.com.

With this latest release, DeepSeek has positioned itself as a significant global player, proving that open models can lead on performance — not just transparency.

Dan Taylor
View More Posts
Dan Taylor is the founder and editor of International Search Insights (ISI), the publication behind ReadIsi, and a Partner and Head of Innovation (Search & AI) at SALT.agency. He has worked in search for more than a decade, in-house and agency-side, and is regarded as one of the West’s leading voices on search engines beyond Google, including Yandex, Baidu, Naver and Cốc Cốc. On ReadIsi, Dan covers the engines and AI answer platforms that English-language coverage tends to miss: Baidu and ERNIE, Yandex and Alice AI, Naver, Cốc Cốc, Shenma, Sogou, WeChat and Weibo search, plus DeepSeek and Alibaba’s Qwen. He also keeps dated records of documented changes, such as the Yandex, Baidu and Naver algorithm update histories and the DeepSeek, ERNIE and Qwen model update histories. His approach is to take learnings from global search engines and apply logic with a high evidence threshold. A claim needs a source, patterns seen in one market are treated as a hypothesis until they are checked in another, and where the evidence is thin he says so. For brands, that means advice built on what can be shown, with measurement that separates being cited from being chosen. Dan is credited with coining the term “edge SEO” and won the inaugural TechSEO Boost research competition in 2018 for his work on CDN and edge changes for SEO. In 2026 he was credited with coining “Infinite Tail” to describe how keyword research is changing with AI search. He is the author of AI Search Marketing For The Travel Sector, writes for Search Engine Journal, Search Engine Land and Travel Marketing Guy, and has advised technology, SaaS, travel and enterprise ecommerce brands, including Cloudflare, GitLab and Airbnb. He has given multiple talks and panels at the International Search Summit in Barcelona, and has also spoken at Optimization Moscow, brightonSEO, Google Search Central Live Deep Dive Asia in Bangkok, Athens SEO and the Asia SEO Virtual Summit 2026, where his talk looked at how Baidu, Naver, DeepSeek, Tencent, Cốc Cốc and Yandex built their own AI answers.
Founder & Editor