OpenAI unveiled GPT-4o in May 2024 – improving on the GPT-4 Turbo model and solving the im-a-good-gpt2-chatbot mystery. It’s not GPT-5 but OpenAI's new flagship model is impressive – GPT-4o can reason across audio, vision, and text in real time.
If you’ve been disappointed by LLMs being able to complete useful tasks so far, it’s time to try again. GPT-4o combines text, vision (images), and audio in a whole new way. For example if you need to solve a math problem, fix a problem in code, or translate a foreign language in real time.
It’s faster, cheaper, and more human than GPT-4 Turbo (and other leading models). GPT-4o has a 128K context window and has a knowledge cut-off date of October 2023. GPT-4o is 50% cheaper than GPT-4 Turbo, across both input tokens ($5 per million) and output tokens ($15 per million).
All ChatGPT users have access to GPT-4o – free users just have limited uses before they have to upgrade to a paid plan.
Here’s what GPT-4o can do so far.
GPT-4o is more human than other AI models
One of the most impressive things is how much more human AI voice interactions sound with GPT-4o in real time.
And not only does it sound more human, it helps with very specific human tasks like tutoring math, and real-time translation of foreign languages.
Watch GPT-4o teach math to a student
Watch GPT-4o perform real-time translation
It’s obvious that OpenAI got the focus on human interaction down. And while it’s not the 10x multiplier GPT-5 is expected to be, GPT-4o is smarter than all other models in major performance benchmarks (except one.)
GPT-4o performance compared to other LLMs
GPT-4o beats all the other major LLMs in most performance categories. Claude 3 Opus does slightly better with Multilingual Grade School Math (MGSM.) Otherwise, it edges out all the top categories.
To anyone using it, it’s easy to see GPT-4o is much faster than GPT-4.
That’s largely because GPT-4o is 2x faster at generating tokens than GPT-4 Turbo. And it’s proven to be adept at advanced tasks. It’s advantages will become even more clear as developers start building things with the GPT-4o API.
GPT-4o API for developers
Developers can now access GPT-4o in the API as a text and vision model. It’s 2x faster, half the price, and has 5x higher rate limits than GPT-4 Turbo. The new audio and video capabilities are coming to a small group of trusted API partners soon.
Here’s the list of GPT-4o API features:
- GPT-4 Turbo-level performance on text, reasoning, and coding intelligence, while setting new high watermarks on multilingual, audio, and vision capabilities
- GPT-4o is 50% cheaper than GPT-4 Turbo, across both input tokens ($5 per million) and output tokens ($15 per million)
- GPT-4o will have 5x the rate limits of GPT-4 Turbo—up to 10 million tokens per minute. (We’re ramping up rate limits to this level in the coming weeks for developers with high usage)
- GPT-4o has improved vision capabilities across the majority of tasks
- GPT-4o has improved capabilities in non-English languages and uses a new tokenizer which tokenizes non-English text more efficiently than GPT-4 Turbo
While developers can get to work on the API, ChatGPT users around the world just got a huge upgrade to their free and plus subscriptions.
GPT-4o in ChatGPT
Instead of using the slower, not-as-smart GPT-3.5, ChatGPT users will all be upgraded to GPT-4o access. Free ChatGPT users will get a lot of the power from GPT-4o.
Free ChatGPT user upgrades:
- Experience GPT-4 level intelligence
- Get responses from both the model and the web
- Analyze data and create charts
- Chat about photos you take
- Upload files for assistance summarizing, writing or analyzing
- Discover and use GPTs and the GPT Store
- Build a more helpful experience with Memory
- More …
But, there will be a limited number of messages that free users can send with GPT-4o (depending on usage and demand). When you reach the limit, you’ll automatically switch back to GPT-3.5.
If you’re already paying for ChatGPT Plus or Team, don’t worry, you get better access to GPT-4o than free accounts.
ChatGPT Plus users get:
- Up to 5x more messages
- More reliable access during peak times
- Early access to features like the new Voice Mode ("Her")
- Quicker model responses
- New capabilities, including audio and video AI creator features
While everyone figures out how to use GPT-4o, businesses are busy integrating it into their apps, tools, and processes. If you have the right tech stack and team in place, switching to the faster, cheaper, and smarter GPT-4o is easy.
How do I hire senior AI engineers to build with GPT-4o?
You could spend the next 6-18 months planning to recruit and build an AI team (if you can afford it), but in that time, you won’t be building any AI capabilities. That’s why Codingscape exists.
We can assemble a senior AI development team for you in 4-6 weeks. It’ll be faster to get started, more cost-efficient than internal hiring, and we’ll deliver high-quality results quickly.
Zappos, Twilio, and Veho are just a few companies that trust us to build their software and systems with a remote-first approach.
You can schedule a time to talk with us here. No hassle, no expectations, just answers.
