{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiduvxviimhae25snu2yutrpikt2qnyrjia5ubao65j37k4xwsziju",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mjmcle7bs6a2"
  },
  "path": "/t/dedicated-cpu-inference-endpoint-returns-empty-http-500-after-80s-is-there-a-configurable-request-timeout/175278#post_3",
  "publishedAt": "2026-04-16T11:09:20.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "Thanks for the detailed investigation! Based on your findings, here are a few things to check that might resolve the 80s timeout:\n\n  1. **Adjust Request Timeout Settings** :\n\n     * In the `huggingface_hub` library, ensure the `timeout` parameter is set higher than 80s when initializing the `InferenceApi` client.\n     * Example: `InferenceApi(repo_id=..., timeout=120)`\n  2. **Verify Container Resource Limits** :\n\n     * Although memory usage is low, confirm if the CPU cores allocated are sufficient for your workload. Sometimes CPU throttling can cause unexpected halts.\n  3. **Check Server-side Logs** :\n\n     * If possible, enable DEBUG level logs on the endpoint side to see if there is a silent exception being caught that isn’t visible in the standard 500 error message.\n\n\n\nHope this helps fix the empty 500 response issue!",
  "title": "Dedicated CPU Inference Endpoint returns empty HTTP 500 after ~80s: is there a configurable request timeout?"
}