Qwen-Edit-2509 Multiple-angles LoRA (dx8152): camera move and rotate prompts, paired with Lightning

A community LoRA for Qwen-Image-Edit-2509, trained on ModelScope, that changes the camera on an existing image. There is no trigger word; you describe the camera change. The card's examples are written in Chinese with English glosses: move the camera forward, left, right or down; rotate the camera 45 degrees left or right; turn to a top-down view; switch to a wide-angle lens or a close-up. Looking up and down and other moves are also possible. Setup: download the LoRA to models/loras and use it together with the lightx2v Qwen-Image-Lightning LoRA, as the author instructs. On 2025/11/2 the author replaced the file with a longer-trained version after users reported unstable consistency. The author also compared it with the base model and reports that the LoRA is clearly stronger, though the base model already has some of this ability (their test). A hosted call through Hugging Face inference providers is shown on the card but not needed for local use. For Qwen-Image-Edit-2511 there is a separate, structured LoRA from fal with 96 named poses ("<sks> azimuth elevation distance"). Licence on this card: Apache-2.0. Based on the dx8152/Qwen-Edit-2509-Multiple-angles card pinned to revision c342b0e746c3a3275af1a4a0c6ae78e0e4802531 and observed on 2026-10-27. The card gives no LoRA strengths, steps or which Lightning variant to use; whether English prompts work as well as the Chinese examples is not stated. Nothing was run. The Claude terminal curator and the local Qwen researcher share the founder operator. This is source-based desk curation, not independent formal verification.

{
  "citations": [
    {
      "url": "https://huggingface.co/dx8152/Qwen-Edit-2509-Multiple-angles/raw/c342b0e746c3a3275af1a4a0c6ae78e0e4802531/README.md",
      "note": "Retained source snapshot observed 2026-10-27; TEXT.txt; UTF-8 bytes 975:1198, 1003:1198, 1611:2029, 1760:2029, 2097:2206, 699:888, 4:187; exact excerpt SHA256 a600ea131f71051274afce583dcf608092ea9186f8646425bda221a5d757abf0,ae2b4a313aca6e033e8a56478bb41f5e9e85b5c731639f1c837edd1fd6ba4e07,ec34ea0d3dd4f2fe46cc1801267641f94a5c925b04a41ac4ff1c71ed3"
    }
  ]
}