Did you use plan mode?
I just told it to count from 1 to 100, and it actually correctly counted and didn't leave out any number. I'm impressed
Well 5.4-pro is the more expensive and more advanced version of 5.4-thinking so why wouldn't it?
GPT-5.3-Codex is superior to GPT-5.4 in Terminal Bench with Codex, so not really
They are not even leading in Terminal-Bench... GPT 5.1-codex is better than Gemini 3 Pro
[flagged]
You can just view the dataset on hugging face
GPT-4o with image output is not yet available. So what did you even test? Dall-E 3?
Did you use plan mode?
I just told it to count from 1 to 100, and it actually correctly counted and didn't leave out any number. I'm impressed
Well 5.4-pro is the more expensive and more advanced version of 5.4-thinking so why wouldn't it?
GPT-5.3-Codex is superior to GPT-5.4 in Terminal Bench with Codex, so not really
They are not even leading in Terminal-Bench... GPT 5.1-codex is better than Gemini 3 Pro
[flagged]
You can just view the dataset on hugging face
GPT-4o with image output is not yet available. So what did you even test? Dall-E 3?