<h1 id="z-image-turbo_clear">Z-Image-Turbo_clear</h1><ul><li>Guide (External site): <a target="_blank" rel="ugc" href="https://www.ai-image-journey.com/2025/12/z-image-turbo-clear.html">English</a> | <a target="_blank" rel="ugc" href="https://note.com/ai_image_journey/n/nd1323246e146">Japanese</a></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/easygoing0114/Z-Image-Turbo_clear">Hugging Face</a></li><li>VAE: <a target="_blank" rel="ugc" href="https://civarchive.com/models/2191617/z-imageclearvae">Z-Image_clear_vae</a></li></ul><h2 id="compare-to-z-image-turbo-(original)">Compare to Z-Image-Turbo (Original)</h2>Z-Image-Turbo_clear | Z-Image-Turbo<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e24c06f8-2b7a-4011-a69e-87a181524f51/width=525/e24c06f8-2b7a-4011-a69e-87a181524f51.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0a7fa685-7aae-4c4b-92f4-9239e686f1e6/width=525/0a7fa685-7aae-4c4b-92f4-9239e686f1e6.jpeg" />The Z-Image-Turbo_clear model is a fine-tuned version of Z-Image-Turbo that improves details compared to the original.For better results, please use it with the dedicated VAE, <a target="_blank" rel="ugc" href="https://civarchive.com/models/2191617/z-imageclearvae">Z-Image_clear_vae</a>.<hr /><h2 id="which-precision-should-i-use">Which Precision Should I Use?</h2><pre><code>+----------+------+----------+----------+--------------------+-----------+
| Format | Sign | Exponent | Mantissa | Significant Digits | Precision |
+----------+------+----------+----------+--------------------+-----------+
| FP32 | 1bit | 8bit | 23bit | 7–8 digits | ✅ |
| BF16 | 1bit | 8bit | 7bit | 2–3 digits | 👍 |
| FP16 | 1bit | 5bit | 10bit | 3–4 digits | 👍 |
| FP8(e5m2)| 1bit | 5bit | 2bit | 1–2 digits | ⚠️ |
| FP8(e4m3)| 1bit | 4bit | 3bit | 1–2 digits | ⚠️ |
+----------+------+----------+----------+--------------------+-----------+</code></pre><ul><li>Exponent: Stability</li><li>Mantissa: Quality</li></ul><h3 id="anime-illustration-comparison">Anime Illustration Comparison</h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c1390b8e-fd34-4d61-ac2e-52b6c2c28da4/width=525/c1390b8e-fd34-4d61-ac2e-52b6c2c28da4.jpeg" /><h3 id="hand-close-up">Hand Close-up</h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/318a8fc5-1ff3-4bae-81fe-31a1eac8c6c1/width=525/318a8fc5-1ff3-4bae-81fe-31a1eac8c6c1.jpeg" /><ul><li>FP32: Highest precision, high VRAM usage</li><li>BF16: Most popular in Z-Image-Turbo model, good for LoRA compatibility</li><li>FP16: Slightly better precision than BF16, less performance degradation on pre-RTX 2000 or non-NVIDIA GPUs</li><li>FP8(e4m3): Lower precision, low VRAM usage, accelerated on RTX 4000 series and later</li></ul>If you are unsure which precision to use, it is safest to choose the BF16 format.<hr /><h2 id="recomended-settings">Recomended Settings</h2><ul><li>Sampler:<ul><li>euler</li><li>res_multistep</li></ul></li><li>Scheduler: beta</li><li>Steps: 9</li><li>ModelSamplingAuraFlow: shift 7.5 (3-9)</li></ul><hr /><h2 id="upscaler">Upscaler</h2>your choice, examples:<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/easygoing0114/AI_upscalers/tree/main">4x-SwinIR-L_GAN</a></li><li><a target="_blank" rel="ugc" href="https://openmodeldb.info/models/4x-NMKD-Superscale">4x_NMKD-Superscale-SP_178000_G</a></li></ul><hr /><h2 id="tips-(external-sites)">Tips (External Sites)</h2><ul><li>ComfyUI VRAM management: <a target="_blank" rel="ugc" href="https://www.ai-image-journey.com/2025/03/comfyui-setting.html">English</a> | <a target="_blank" rel="ugc" href="https://note.com/ai_image_journey/n/nc46cfff6c3db">Japanese</a></li><li>Steps and Scheduler: <a target="_blank" rel="ugc" href="https://www.ai-image-journey.com/2025/06/step-schedulers.html">English</a> | <a target="_blank" rel="ugc" href="https://note.com/ai_image_journey/n/n0878f3eb6e6e">Japanese</a></li><li>Color Correct custom nodes in ComfyUI: <a target="_blank" rel="ugc" href="https://www.ai-image-journey.com/2025/09/comfyui-easygoing-nodes.html">English</a> | <a target="_blank" rel="ugc" href="https://note.com/ai_image_journey/n/n5bb33311b866">Japanese</a></li></ul><hr /><h2 id="license">License</h2><ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/easygoing0114/Z-Image-Turbo_clear/blob/main/LICENSE">Apache-2.0</a></li></ul>

Z-Image-Turbo_clear

zImageTurboClear_fp8SvdScaled.safetensors

Files with SHA256: 993918b5a63dba8ca292017bf6175685ad171acf0f1fff7ab7e2d0d396037b84