Ask HN: Codex is too slow. Is there any solution?

6 pointsrule20255mo ago6 comments

The codex backend is of good quality, the frontend is average, but most importantly it is too slow. I wonder if OpenAI will improve it.

6 comments

moomoo115mo ago

It seems to work with less issues than CC opus.

I don’t mind if it takes longer as long as the answer is correct more often.

You can always be doing more work while one chat is working..

naiv5mo ago

the new 0.47 has a better performance now imho

i_have_an_idea5mo ago

this seems like a crazy idea as the cli client has nothing to do with how many tokens per second the api streams

naiv5mo ago

The api is never the bottleneck but how fast the cli provides context. So just by using ripgrep it will be faster than using grep. On top of this concurrent code search compared to sync etc

esafak5mo ago

Sonnet and Gemini are good and fast. Can't speak for Grok.

muzani5mo ago

Grok Turbo is fast.

j / k navigate · click thread line to collapse

Ask HN: Codex is too slow. Is there any solution?

6 pointsrule20255mo ago6 comments

The codex backend is of good quality, the frontend is average, but most importantly it is too slow. I wonder if OpenAI will improve it.

6 comments

moomoo115mo ago

It seems to work with less issues than CC opus.

I don’t mind if it takes longer as long as the answer is correct more often.

You can always be doing more work while one chat is working..

naiv5mo ago

the new 0.47 has a better performance now imho

i_have_an_idea5mo ago

this seems like a crazy idea as the cli client has nothing to do with how many tokens per second the api streams

naiv5mo ago

The api is never the bottleneck but how fast the cli provides context. So just by using ripgrep it will be faster than using grep. On top of this concurrent code search compared to sync etc

esafak5mo ago

Sonnet and Gemini are good and fast. Can't speak for Grok.

muzani5mo ago

Grok Turbo is fast.

j / k navigate · click thread line to collapse