GPT-4o Statistics and Facts

We collected some stats and facts about OpenAIs latest model: GPT-4o. Learn about its capabilities and enhanced speed compared to previous models.

Written by
Daniel Højris Bæk
Calendar Icon - Dark X Webflow Template
May 15, 2024

OpenAI launched its latest model, GPT-4o, on May 13th, 2024.

You can find their announcement page here.

In general, it has already gathered a lot of attention, most of it centered around the multi-model usage, where it uses voice as feedback on what is captured with the phone's camera.

The impressive demonstration of the multi-modal capabilities of GPT-4o

We have collected some facts and stats. And, of course, we also already implemented it inside our SEO.AI platform.

The key takeaway from our own use is that it is approximately 100-120% faster than the GPT-4 Turbo and even slightly faster than the earlier fastest model, the GPT3.5 Turbo.

Key Facts about GPT-4o

  • Multimodal Capabilities: GPT-4o can process and generate text, audio, image, and video inputs and outputs.
  • Response Time: It can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds.
  • Performance: Matches GPT-4 Turbo performance on text in English and code, with significant improvements in non-English languages.
  • Cost and Speed: GPT-4o is 50% cheaper and much faster in the API than the previous top model (GPT-4 Turbo).
  • Model Integration: Combines text, vision, and audio processing in a single neural network.
  • Enhanced Understanding: Better at vision and audio understanding compared to existing models.
  • Evaluation Scores:
    • Achieves 88.7% on 0-shot COT MMLU for general knowledge questions.
    • Sets new high scores on multilingual, audio, and vision capabilities.
    • Outperforms Whisper-v3 in speech recognition and translation.
  • Language Tokenization: Improved token compression across 20 languages, reducing the number of tokens needed.
  • Safety: Built-in safety features, including filtering training data and refining model behavior post-training.
  • External Testing: Extensively tested with over 70 external experts in various domains.
  • Availability:
    • Rolling out text and image capabilities in ChatGPT.
    • Available in the free tier and to Plus users with higher message limits.
    • Developers can access GPT-4o in the API. Audio and video capabilities will soon be launched to trusted partners.

If you are using ChatGPT for SEO (you should really consider SEO.AI to get a better workflow and SEO insights) I did a short review for using GPT-4o for SEO.

Want to try the #1 AI Writer for SEO Copywriting?

Create anything from blog posts to product descriptions with 1-click AI drafts or our chat assistant. Powered by a next-gen SEO engine that ensures your content actually ranks. Try it now with a free trial→

GPT-4o Statistics and Facts

This is an article written by:

+20 years of experience from various digital agencies. Passionate about AI (artificial intelligence) and the superpowers it can unlock. I had my first experience with SEO back in 2001, working at a Danish web agency.