Interpreter mode for gpt-4o

Could you please enable GPT-4o for Interpreter mode? The model has consistently handled all tasks exceptionally well, including highly challenging ones. I am confident that it will be the optimal choice for Interpreter mode.

2 Likes

Pleeeeeeease :weary:

2 Likes

bumping

+1

GPT-4o-2024-05-13 surpassed every other LLM on https://chat.lmsys.org/?leaderboard

This would be a huge improvement.

It really would be an amazing interpreter I don’t know how they haven’t enabled it yet.

In our initial tests, it performed a decent bit worse on interpreter mode/agent loops. We’re not fully convinced of it’s dominance over gpt-4-turbo just yet.

3 Likes

I’d still like to use it. I know it performs well when the context is correct, and your tests shouldn’t limit its use based on your tests. People have the option to choose their own models.

is it true that gpt4o are only better than gpt4-turbo at small tasks that are normally encountered at ChatGPT?

I believe the effectiveness of a model depends on how it is used, the context provided, and the expected outcomes. I assigned an extremely challenging task to GPT-4o, which even a combination of GPT-4 Turbo and Claude Opus struggled with over several days. GPT-4o accomplished it in just two shots.

3 Likes

Could you please add it?

1 Like

bumping