What do you think only GPT-4o can do but GPT-4 cannot?

The GPT4-o is just an optimized version of GPT-4 Turbo. The omni version is likely to have a smalller context window so you won’t be able to do very large prompts or process large documents. The older model handles up to 128K token context. From API point of view, new omni version is: 👉 50% lower pricing 👉 GPT-4o is 50% cheaper than GPT-4 Turbo 👉 2x faster than GPT-4 Turbo 👉 5x higher rate limits