Hugging Face API timeouts on all models!

Inference Endpoints on the Hub
1

:double_exclamation_mark: I’m experiencing multiple timeouts on all models on the API. I’m currently using the free tier. Is anyone else experiencing this issue? is using the free tier part of the issue?

status: 504,
statusText: ‘Gateway Time-out’,
1 Like
2

Hmm…? It works for me.

import os
from huggingface_hub import InferenceClient

HF_TOKEN = "hf_***my_read_token***"

client = InferenceClient(
    provider="hf-inference",
    api_key=HF_TOKEN,
)

completion = client.chat.completions.create(
    model="HuggingFaceTB/SmolLM3-3B",
    messages=[
        {
            "role": "user",
            "content": "What is the capital of France?"
        }
    ],
)

print(completion.choices[0].message)
"""
ChatCompletionOutputMessage(role='assistant', content="<think>\nOkay, the user is asking for the capital of France. Let me make sure I remember correctly. I think it's Paris. Wait, is there any chance they might be confused with another city? Maybe someone might think it's Lyon or Marseille? No, those are major cities but not the capital. The capital of France is definitely Paris. I 
should confirm that. Let me think if there's any historical context that could change this. No, Paris has been the capital since the 10th century. There's also the Eiffel Tower and the Louvre, 
which are famous landmarks there. Yeah, I'm pretty confident it's Paris. I should just state that clearly and maybe add a bit about it being the largest city in France to give more context.\n</think>\n\nThe capital of France is **Paris**. It is not only the largest city in the country but also a major global cultural, economic, and political hub. Paris is renowned for landmarks like 
the Eiffel Tower, the Louvre Museum, and the Seine River, and has been the heart of French history and politics for centuries.", tool_call_id=None, tool_calls=[], refusal=None, annotations=None, audio=None, function_call=None, reasoning_content=None)
"""

No major server issues have been detected so far.

3

Hi @GeorginaSteele Are you still experiencing this issue? Can you please share the model you’re using?

1 Like