Archive:ChatGPT Deep Research: Difference between revisions
Added links |
Added links |
||
| Line 22: | Line 22: | ||
== Performance == | == Performance == | ||
Deep Research scored 26.6% on the "Humanity's Last Exam" benchmark, surpassing rivals like DeepSeek's R1 (9.1%) and GPT-4o (3.3%)<ref name="Entrepreneur"/>. It generates reports exceeding 10,000 words in 5–30 minutes<ref name="BI"/>. | Deep Research scored 26.6% on the "[[Humanity's Last Exam]]" benchmark, surpassing rivals like [[DeepSeek|DeepSeek's]] R1 (9.1%) and GPT-4o (3.3%)<ref name="Entrepreneur"/>. It generates reports exceeding 10,000 words in 5–30 minutes<ref name="BI"/>. | ||
== Applications == | == Applications == | ||