# Constant 503 error for several days when running LLAMA 3.1 **URL:** \ **Category:** Inference Endpoints on the Hub\ **Created:** [September 3, 2024, 5:18pm UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144 "2024-09-03T17:18:00Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![saramagdeline](https://avatars.discourse-cdn.com/v4/letter/s/bc79bd/32.png) [@saramagdeline](https://discuss.huggingface.co/u/saramagdeline)\ **Post date:** [September 3, 2024, 5:18pm UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/1 "2024-09-03T17:18:00Z")
Hi all. Earlier this month I wrote a script incorporating LLAMA 3.1, and it worked well (with timeouts, etc), but for the past week whenever I try to access LLAMA 3.1 at all I get a **503 error** (including for short simple requests, like HF’s sample code). Any idea what the problem may be? I do indeed have Pro. ---
**Author:** ![ChrisCleaner](https://avatars.discourse-cdn.com/v4/letter/c/54ee81/32.png) [@ChrisCleaner](https://discuss.huggingface.co/u/ChrisCleaner)\ **Post date:** [April 19, 2025, 11:16am UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/2 "2025-04-19T11:16:29Z")
Did you find a solution to this problem? I am experiencing the same problem on and off for a couple of days now. ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [April 19, 2025, 11:28am UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/3 "2025-04-19T11:28:35Z")
> [@Getting "502 Server Error: Bad Gateway for url: https://api-inference.huggingface.co/models/meta-llama/Llama-3.2-3B-Instruct" error](https://discuss.huggingface.co/t/getting-502-server-error-bad-gateway-for-url-https-api-inference-huggingface-co-models-meta-llama-llama-3-2-3b-instruct-error/150740): > > i am accessing llama3.2\_3b instruct model via huggingfacehub. while calling llm.invoke it hangs. Everything was working fine before yesterday ie 15.04.25. Is anyone else also facing this issue. When will this be resolved There were at least three reports of the same situation on Discord, so I think it’s probably some kind of error. @michellehbn ---
**Author:** ![gikebe](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/gikebe/32/46247_2.png) [@gikebe](https://discuss.huggingface.co/u/gikebe)\ **Post date:** [April 24, 2025, 11:23am UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/4 "2025-04-24T11:23:48Z")
Am still getting a 503 when running Llama-3.2-3B-Instruct. Any known solution? `huggingface_hub.errors.HfHubHTTPError: 503 Server Error: Service Temporarily Unavailable for url: https://router.huggingface.co/hf-inference/models/meta-llama/Llama-3.2-3B-Instruct` ---
**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [April 24, 2025, 12:43pm UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/5 "2025-04-24T12:43:03Z")
Depending on the model, it seems that this can be avoided by the following methods. > [@Inference API stopped working](https://discuss.huggingface.co/t/inference-api-stopped-working/150492/22): > > Mine stopped for a while too getting 503: “503 Server Error: Service Temporarily Unavailable for url: [https://router.huggingface.co/hf-inference/models/black-forest-labs/FLUX.1-schnell](https://router.huggingface.co/hf-inference/models/black-forest-labs/FLUX.1-schnell)” I looked that you could set a provider. I changed the parameters to the following and it worked. from huggingface\_hub import InferenceClient client = InferenceClient( provider="fal-ai", api\_key="hf\_xxxxxxxxxxxxxxxxxxxxxxxx", ) Which you can find if you click view code on the model card. Looks like … ---
**Author:** ![gikebe](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/gikebe/32/46247_2.png) [@gikebe](https://discuss.huggingface.co/u/gikebe)\ **Post date:** [April 25, 2025, 6:00am UTC](https://discuss.huggingface.co/t/constant-503-error-for-several-days-when-running-llama-3-1/105144/6 "2025-04-25T06:00:12Z")
Thanks. Using providers works like magic. Here is the documentation: [HF Inference](https://huggingface.co/docs/inference-providers/en/providers/hf-inference) Sample ```auto client = InferenceClient( provider="hf-inference", token=os.environ["HF_TOKEN"]) ```