When LoRA addons fail to appear in Stable Diffusion, it can be frustrating for both beginners experimenting with custom styles and advanced users managing complex model libraries. LoRA (Low-Rank Adaptation) files are lightweight and powerful, but they depend on correct folder placement, naming conventions, UI compatibility, and system configuration. Fortunately, most visibility issues are caused by common setup mistakes that can be fixed within minutes.
TLDR: If LoRA addons are not showing in Stable Diffusion, the issue usually relates to incorrect folder placement, unsupported file formats, outdated WebUI versions, missing model refresh, or naming errors. Beginners should first verify file paths and restart the interface. Advanced users may need to check console logs, extension conflicts, or compatibility with specific checkpoints. Following the six troubleshooting steps below resolves nearly all LoRA visibility problems.
Understanding How LoRA Works in Stable Diffusion
LoRA files are small add-on models designed to modify or enhance a base checkpoint without replacing it. Instead of training a full model from scratch, LoRA injects learned style or concept data into an existing Stable Diffusion model at generation time. This makes them efficient and easy to swap in and out.
However, for Stable Diffusion WebUI (such as AUTOMATIC1111 or similar interfaces) to detect them, the files must be placed in the correct directory and properly formatted. If even one of these requirements fails, the LoRA will not appear in the selection panel.
Image not found in postmetaTroubleshooting Step 1: Verify the Correct Folder Location
The most common issue beginners encounter is placing the LoRA file in the wrong directory.
In most standard installations, LoRA files should be located in:
- stable-diffusion-webui/models/Lora
Common mistakes include:
- Placing the file in the Stable-diffusion folder instead of Lora
- Accidentally creating a nested folder (e.g., Lora/Lora)
- Uploading the file into an extension-specific directory
Solution: Move the LoRA file directly into the correct Lora directory and restart the WebUI. Stable Diffusion does not always update dynamically without restarting.
Troubleshooting Step 2: Check File Format and Extraction
LoRA files typically use the following formats:
- .safetensors (recommended)
- .ckpt (older format)
If the downloaded file is:
- A ZIP or RAR archive
- A file with an unusual extension
- A partially downloaded file
It will not appear in Stable Diffusion.
Solution:
- Ensure the archive is fully extracted.
- Confirm the file extension is correct.
- Redownload the file if its size seems suspiciously small.
Advanced users should also verify file integrity if downloaded from third-party hosting sources.
Troubleshooting Step 3: Refresh or Restart the WebUI
Stable Diffusion does not always auto-detect new LoRA files immediately. Even experienced users occasionally forget this step.
Depending on the interface being used, try:
- Clicking the refresh button next to the LoRA dropdown
- Fully stopping the WebUI process
- Restarting the launch script (e.g., webui-user.bat)
Pro tip for advanced users: Watch the terminal window while restarting. If the LoRA loads correctly, it usually appears in the startup log.
Troubleshooting Step 4: Confirm Base Model Compatibility
Not all LoRA addons work with every base model. LoRAs are often trained for:
- SD 1.5
- SDXL
- Custom checkpoints
If a LoRA is trained specifically for SDXL and the user loads SD 1.5, it may not appear or may cause generation errors.
Solution:
- Check the LoRA description page for compatibility details.
- Match the LoRA version to the active checkpoint.
- Switch models and refresh the LoRA list.
Advanced users working with merged checkpoints should ensure that the LoRA’s training base matches the merge architecture.
Troubleshooting Step 5: Inspect Naming Conventions and Special Characters
Sometimes the issue is surprisingly simple: the file name.
Stable Diffusion may fail to register LoRA files with:
- Excessively long file names
- Special characters such as #, %, &, or non standard symbols
- Leading or trailing spaces
Solution: Rename the file to something clean and simple, such as:
- animeStyle_v1.safetensors
- portraitEnhancer.safetensors
This small change often resolves detection problems instantly.
Troubleshooting Step 6: Check for Extension Conflicts or Outdated WebUI
Advanced users frequently install multiple extensions to enhance functionality. However, outdated plugins or UI modifications can interfere with LoRA detection.
Common advanced-level causes include:
- Outdated AUTOMATIC1111 version
- Broken additional networks extension
- Conflicts between SDXL handling extensions
- VRAM optimization scripts interfering with scanning
Solution:
- Update the WebUI using git pull.
- Disable extensions temporarily to test conflicts.
- Clear the browser cache and reload.
- Check terminal logs for LoRA loading errors.
Power users can also run Stable Diffusion with minimal extensions enabled to isolate the root cause.
Additional Advanced Diagnostics
If none of the above steps work, consider deeper system analysis:
- File permissions: Ensure the user has read access to the LoRA folder.
- Disk path length limits: Windows systems sometimes fail with long directory paths.
- GPU memory initialization errors: Rare, but visible in console logs.
- Corrupt cache files: Deleting temporary cache folders can help.
Reviewing launch logs carefully often reveals the exact problem. Stable Diffusion usually reports when a model fails to load due to compatibility or formatting issues.
Beginner vs Advanced Fix Priority
For beginners, follow this order:
- Check folder location
- Verify file format
- Restart WebUI
For advanced users, focus on:
- Model compatibility
- Extension conflicts
- Log file diagnostics
In practice, over 80% of “LoRA not showing” cases are resolved within the first three steps.
Frequently Asked Questions (FAQ)
Why does my LoRA work on one model but not another?
LoRAs are trained on specific base architectures such as SD 1.5 or SDXL. If the active checkpoint differs from the training base, the LoRA may not appear or may produce distorted results.
Do I need an internet connection for LoRAs to load?
No. LoRAs are local files. Once downloaded and placed correctly, they function offline.
Can too many LoRA files slow down the interface?
Yes. Large libraries can increase scanning time during startup. Organizing rarely used LoRAs in subfolders can improve performance.
Why does the LoRA appear but cause errors during generation?
This often indicates compatibility issues with the base model or incorrect weight values in the prompt.
Is .safetensors better than .ckpt?
Yes. The .safetensors format is generally safer and less prone to corruption, and it is widely recommended.
Do I need a powerful GPU to use LoRAs?
No additional GPU power is required beyond what Stable Diffusion already needs. LoRAs are lightweight compared to full checkpoints.
What if none of the steps fix the issue?
Review console logs carefully, update the WebUI, and test with all extensions disabled. In rare cases, a clean reinstall of Stable Diffusion may resolve persistent configuration corruption.
By systematically applying these six troubleshooting steps, users at all experience levels can quickly restore missing LoRA addons and return to generating images without disruption. In most situations, the solution is surprisingly simple—and once fixed, future installations become second nature.
