[HN Gopher] Show HN: Unofficial OpenAI Status Dashboard
___________________________________________________________________
Show HN: Unofficial OpenAI Status Dashboard
Whenever I get abnormally slow results or high rates of errors from
GPT-4, I used to go and check the official OpenAI status site. But
mostly it would always show green when I was experiencing issues.
So Eliot who I work with built this little utility - an unofficial
OpenAI status dashboard. It's designed to answer the question of
"is GPT slow for everyone right now or just for me" and "is GPT
having elevated rates of errors for everyone right now or just for
me" It's aimed at API users but may also be useful for ChatGPT
errors/slowness, I'm not sure. Possible features we're thinking
about: * Gauge style chart to be easier to read * Show the
average and delta vs average in the chart * Track a specific
prompt over time to track any model performance degradation * Add
stats specifically for ChatGPT (any ideas on how to do so, given
that it's behind Cloudflare?) * Add speed comparison of APIs via
Azure vs OpenAI Let us know what's helpful about it and what you'd
hope to see improved or added.
Author : tikkun
Score : 8 points
Date : 2023-07-13 15:07 UTC (7 hours ago)
(HTM) web link (openai-status.llm-utils.org)
(TXT) w3m dump (openai-status.llm-utils.org)
| haha69 wrote:
| Can you explain where you get this data from? You have it
| scheduled to poll every hour? How can you be sure the data is
| statistically significant? etc
| tikkun wrote:
| Polling every ten minutes, I don't think the data is
| statistically significant, but it is still helpful for me as a
| user
| bitterblotter wrote:
| Pretty cool stuff. Way more concrete that the actual site when
| you measure a proper metric like time to completion.
|
| The charts look too dense and jagged on mobile though.
|
| Also this could be pretty smart to have if the speed is more
| important than the model. Then you could automatically select the
| fastest model from a prioritized list of models, in case the
| better ones become slow.
| obiefernandez wrote:
| Can I get a textual representation of this in a format that would
| be easy to digest by my GPT-4 backed Customer Success Manager bot
| on https://olympia.chat
| bitterblotter wrote:
| Wouldn't it defeat the purpose if your chat bot itself is
| running the service that is slow?
|
| Either way it would probably be easier for you to create the
| textual representation yourself, since it would require plenty
| of tweaking in order to get the result you want.
|
| Maybe you could create pairs of standard deviations of the
| current speed and text, like this:
|
| -2: very slow, -1: slow, 0: working fine, 1: fast, 2: very fast
|
| Then plug the rounded stddev of the current speed into the
| dictionary and insert it into the sentence: "The chat bot is X
| right now"
___________________________________________________________________
(page generated 2023-07-13 23:02 UTC)