Sign In
homemodelsimagesvideos3D Modelsarticlescomicsbountieschallengesupdatesshop

Uncensored Qwen3 4B (LLM for Z-Image & Flux4B)

Updated: May 26, 2026

base modelz-image-turboqwen3_4_bqwen3humante

Download

1 variant available

fp8 SafeTensor

model.safetensors

8-bit, smaller file • 4.83 GB

Verified:

Type
Checkpoint Trained
Stats

4,092

Reviews

(74)

Published

May 26, 2026

Base Model

Flux.2 Klein 4B

Training
Steps: 50,000
Epochs: 50
Hash
AutoV2
418F1E9661
default creator card background decoration
Reactions - 62282

62.3K

Downloads - 580300

580.3K

Generations - 4872731

4.9M

SDXL Training Contest Participant

License:

Apache 2.0
FP8_00005_.png

Qwen_3_4B Trained Text Encoder

For use in Z-Image or FLUX Klein4B

Without lora's LLM had similar text errors to BF16 base QWEN, but better adherence to nude prompt.

See the comparison article here