Skip to content
TheChatArchiveThe Chat Archive
Perplexitysonar-reasoning· PerplexityHugging Face LMArena Search Arena 24K

deepseek v3 0324 gpqa benchmark? Answer in extremely short.

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with gpt-4o-search-preview-high’s answer.
  1. User

    deepseek v3 0324 gpqa benchmark? Answer in extremely short.

  2. Perplexity Assistant

    DeepSeek-V3-0324 achieved 68.4 on the GPQA benchmark, a +9.3 improvement from its previous version[1][5].

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.