{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreigkvwdfumlrqnrkkuoojhpkjnkqezg2o4rbb6v7nroeompxo5jeby",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mi6aky6yc2r2"
  },
  "path": "/t/stable-diffusion-xl-much-slower-with-candle-than-with-diffusers/174649#post_3",
  "publishedAt": "2026-03-28T21:57:02.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "I already tried to change the float type to f16 for the decover, the change is barely noticeable.\n\nAs suggested, I’ve tried patching the scheduler, using the same configuration as found in the `scheduler.json` in the repository of the model, but that changes nothing: the speed is exactly the same as before, and I still get weirdly off looking images that I don’t get with `diffusers`.\n\nI think the next step will be to try to use the `diffusers-rs` library, which seems to be backed by Torch rather than candle.\n\nIncidentally, I tried to use Torch compilation with the `diffusers` library and that resulted in a huge speedup. With it, the diffusion steps can reach the speed of 4 steps per second.",
  "title": "Stable Diffusion XL much slower with candle than with diffusers"
}