{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidfkwn7ddtfq7jj34yzehxsrbctvkhh3as3tn6krkhroyqyuay7ya",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mmphobelvny2"
  },
  "path": "/t/how-to-convert-a-single-safetensors-file-to-peft-format/173103#post_11",
  "publishedAt": "2026-05-25T19:57:59.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "Qwen-Image-2512-Lightning",
    "OpenxAILabs/Qwen-Image-2512-Lightning-8steps-V1.0-bf16-PEFT · Hugging Face"
  ],
  "textContent": "Thank you for this script! I can confirm it indeed allows the lora to be ran in vllm-omni.\n\nI did notice however that the size drops significantly (from 850 MB to 378 MB), is there some information loss when applied to the new Qwen-Image-2512-Lightning?\n\nI’ve uploaded the output: OpenxAILabs/Qwen-Image-2512-Lightning-8steps-V1.0-bf16-PEFT · Hugging Face",
  "title": "How to convert a single safetensors file to PEFT format"
}