Hey, I just want to ask if the LLAMA’s Vision models are down right now. I’ve encountered issues with them via the API and Playground since yesterday or it’s just me.
1 Like
Hello Markpaine.it
I’m not able to reproduce your issue. I did a quick test with jpeg files on the playground for both 11 and 90b. Can you tell me a bit about what you are doing if you are able to reproduce it consistently?
Thanks
Seth Kneeland
@markpaine.it we are getting consistant responses on our end. Can you do me a favor and gather a network diagnostic when the error occurs so I can see what error code you are getting behind the scenes?
You actually exceed the rate limit. See the 429 in the header. I actually have an RFE to the cloud development team to put the real message in the body as well as the header to prevent the perception that our end points are not available.
-Coby
1 Like
Alright thank you for the information.
Hi @coby.adams
I’ve encountered the same error message:
Llama-3.2-11B-Vision-Instruct is temporarily unavailable. Please try again later!
However, I’m convinced it’s not a rate limit issue. Even after waiting for more than 12 hours, the error persists. Moreover, when I do exceed the rate limit, I receive the clear message Rate limit exceeded. This makes the 429 status code quite confusing.
Could you help me figure out what’s going on? Thanks a lot!
1 Like
The only thing I can ask is if you do get that error message again do a browser inspect and lets look at the network cals and see what response we are getting back. Without the header response I cannot triage. There was nothing wrong system wide or with the model last night.
-Coby
tried the vison model just now, it works now.
1 Like