Loading article...

SoloTagger v0.20: Improved Caption Output Quality

  SoloTagger is an LLM-based image captioning tool.  
## What's Changed
1. The prompt passing mechanism has been optimized, significantly improving caption output quality.
2. The default temperature has been changed from 0.1 to 0, reducing model hallucinations.

## Note:
1. The prompt is the most important factor affecting caption output quality. Please refine and adjust it according to your specific needs.
2. In my own experience, with a well-crafted prompt, Qwen3.5-9B-UD-Q5KXL.gguf produces better output quality than llama-joycaption-beta-one-hf-llava.i1-Q6_K.gguf.
3. Among the Qwen3.5 series, models smaller than 9B produce noticeably lower output quality.

SoloTagger v0.20 UI

For detailed usage instructions, please refer to the previous documentation:
v0.12: Easier Prompt Editing
v0.10: Setup on Windows via LM Studio

Download: SoloTagger_v0.20.zip
GitHub: https://github.com/sololo-xyz/SoloTagger

You Might Also Like View All
Cover: SoloCropper: Efficient Human Image Cropping for Dataset Preparation
SoloCropper: Efficient Human Image Cropping for Dataset Preparation
Cover: Gemma 4: A Powerful New Weapon for SoloTagger
Gemma 4: A Powerful New Weapon for SoloTagger
Cover: Z Image LoRA Commissions Now Open
Z Image LoRA Commissions Now Open
Cover: Price Update for Model Orders
Price Update for Model Orders
Cover: SoloTagger v0.12: Easier Prompt Editing
SoloTagger v0.12: Easier Prompt Editing
Cover: SoloTagger: Local JoyCaption Beta One GGUF Setup on Windows via LM Studio
SoloTagger: Local JoyCaption Beta One GGUF Setup on Windows via LM Studio
Cover: Basic Workflow: FLUX.2-klein I2I v2.2  |  4-in-1 image editing
Basic Workflow: FLUX.2-klein I2I v2.2 | 4-in-1 image editing
Cover: Training Notes 2: Z Image LoRA - Phase Recap
Training Notes 2: Z Image LoRA - Phase Recap
Cover: The new version is coming. What surprises will it bring to the world this time?
The new version is coming. What surprises will it bring to the world this time?
Cover: Training Notes: My Take on Klein & Z Image
Training Notes: My Take on Klein & Z Image
Cover: Basic Workflow: Z Image T2I v1.3
Basic Workflow: Z Image T2I v1.3
Cover: Basic Workflow: FLUX.2-klein I2I v2.0  |  4-in-1 image editing
Basic Workflow: FLUX.2-klein I2I v2.0 | 4-in-1 image editing
Table of Contents