Train a Z-Image Turbo LoRA

title: "Z-Image Turbo LoRA: Dataset, Precision, Adapter, AI Toolkit" description: "Train Z-Image Turbo LoRAs with AI Toolkit: dataset sizing, fp8 quantization, training adapter, sigmoid timestep, and Flux/SDXL differences." slug: "train-z-image-turbo-lora" keywords:

• "train z-image turbo lora"
• "z-image turbo training adapter"
• "z-image turbo vram requirements"
• "z-image turbo ai toolkit config"
• "z-image turbo sigmoid timestep"

updated: "2026-07-27" sources:

• "https://huggingface.co/Tongyi-MAI/Z-Image-Turbo"
• "https://huggingface.co/Tongyi-MAI/Z-Image"
• "https://huggingface.co/ostris/zimage_turbo_training_adapter"
• "https://github.com/ostris/ai-toolkit"
• "https://github.com/citronlegacy/ai-toolkit-zimage-config-generator/blob/main/example_configs/zimage_AnubisCitronOC_ZIT_v3.yaml"
• "https://arxiv.org/abs/2511.22699"
• "https://www.runcomfy.com/comfyui-nodes/Comfyui_CXH_joy_caption"
• "https://civitai.com/articles/23233/heres-how-to-quickly-train-a-z-image-turbo-lora"
• "https://civitai.com/articles/23842/z-image-turbo-character-lora-training-guide-easy-how-to-guide"
• "https://civitai.com/articles/23314/z-image-turbo-lora-training-with-ai-toolkit-and-z-image-controlnet-full-tutorial-for-highest-quality"
• "https://civitai.com/articles/24472/z-image-turbo-zit-anime-lora-training-experience"

TL;DR

• Dataset: 9–20 diverse images beat larger uniform sets; captions optional — tags work with "no caveats."
• Precision: fp8 (qfloat8) quantization cuts VRAM ~50% vs bf16; 16 GB cards can train with low_vram: true, 24 GB+ needed for unquantized.
• Adapter: You must load the Ostris de-distillation adapter (ostris/zimage_turbo_training_adapter) or the LoRA will lose its 8-step speed.
• Settings that differ from Flux/SDXL: arch: zimage:turbo, sigmoid timestep (weighted), no negative prompts, natural-language prompts preferred.

What is Z-Image Turbo and why does training differ?

Z-Image Turbo is a 6-billion-parameter single-stream diffusion transformer (S3-DiT) from Tongyi Lab (Alibaba) distilled to 8 inference steps via Decoupled-DMD [^1]. The base Z-Image model supports classifier-free guidance (CFG), negative prompts, and 28–50 steps; Turbo drops all three for speed [^2].

Because the model is already distilled, standard LoRA fine-tuning breaks the distillation: the model "reverts toward requiring more inference steps" [^3]. The fix is a de-distillation training adapter — a pre-trained LoRA that has already learned to undo the distillation. You train your LoRA on top of this adapter, then discard it at inference.

Hardware and precision: what VRAM do you actually need?

The official inference card states Turbo "fits comfortably within 16G VRAM consumer devices" [^1]. Training is heavier; community configs default to fp8 for all GPU tiers [^5]. One guide claims "float8 would be slower" [^4]; configs use qtype: qfloat8 with quantize: true as default for all GPU tiers [^5].

Key distinction from Flux/SDXL: Those models typically train in bf16 or fp16. Z-Image Turbo configs default to fp8 quantization for both transformer and text encoder (quantize_te: true, qtype_te: qfloat8) because the distillation is sensitive to precision drift.

The training adapter: v1 vs v2

The Ostris repository hosts ostris/zimage_turbo_training_adapter with two files [^3]:

• zimage_turbo_training_adapter_v1.safetensors — stable, widely used
• zimage_turbo_training_adapter_v2.safetensors — "in testing" per one guide [^6]; no public changelog exists

In AI Toolkit you reference it via assistant_lora_path: ostris/zimage_turbo_training_adapter/zimage_turbo_training_adapter_v1.safetensors [^5]. If you omit the adapter, the distillation breaks down "very quickly" and "in an unpredictable way" [^3].

Dataset: size, captions, and diversity

Natural-language captions may yield better prompt adherence; one guide notes "Use long, detailed, coherent natural language descriptions... The model responds much better to full sentences and paragraphs" [^8]. For local captioning without upload limits, the ComfyUI JoyCaption extension supports siglip-so400m-patch14-384, Meta-Llama-3.1-8B-bnb-4bit, and joy-caption-pre-alpha [^9].

AI Toolkit settings that actually matter

These are the non-default parameters that community configs and guides agree on [^5][^7][^6]:

Trigger word: Must appear in every sample prompt or "early samples look like the base model and it's hard to tell what's working" [^6].

What differs from Flux and SDXL training

One guide claims "Train it the same way as Illustrious" [^8], but the config differences above are structural — not optional.

Common errors and fixes

FAQ

How much VRAM do I need to train a Z-Image Turbo LoRA?

With fp8 quantization (quantize: true, qtype: qfloat8) and low_vram: true, 12–16 GB works. Unquantized training needs 24 GB+. The official inference requirement is 16 GB [^1].

What is the Z-Image Turbo training adapter and why do I need it?

It's a pre-trained LoRA (ostris/zimage_turbo_training_adapter) that "de-distills" the model. Training your LoRA on top of it preserves the 8-step speed. Without it, "the distillation breaks down very quickly" [^3].

Can I train Z-Image Turbo on tags instead of captions?

Yes. "No caveats. It works." [^8] Natural-language captions may yield better prompt adherence; one guide notes "Use long, detailed, coherent natural language descriptions... The model responds much better to full sentences and paragraphs" [^8].

Why does my Z-Image Turbo LoRA need more inference steps after training?

The distillation broke. This happens when training too long without the adapter, or with a high learning rate. Retrain with the adapter and cap at 3000 steps [^3][^6].

What's the difference between training adapter v1 and v2?

v1 is stable; v2 is "in testing" with no public changelog [^6]. Use v1 unless you're experimenting.

Do negative prompts work with Z-Image Turbo?

No. The model card explicitly lists "Negative Prompting: ❌" for Turbo [^2]. Do not include them in prompts or sample configs.

How many training steps should I use?

3000 is the community standard (~1 hour on RTX 5090) [^6]. The concept typically emerges at 1000–1500 steps [^6]. No published data on diminishing returns beyond 3000.

Can I use the RunPod Ostris template for training?

Yes. The official AI Toolkit template is available on RunPod [^6]. Select it, choose your GPU (RTX 5090 used in published runs), and deploy.

Sources

1. Z-Image Turbo model card — architecture, distillation, VRAM, CFG/negative prompt support — https://huggingface.co/Tongyi-MAI/Z-Image-Turbo
2. Z-Image base model card — comparison table (CFG, steps, negative prompting, diversity) — https://huggingface.co/Tongyi-MAI/Z-Image
3. Ostris training adapter card — de-distillation mechanism, distillation breakdown warning — https://huggingface.co/ostris/zimage_turbo_training_adapter
4. Ostris AI Toolkit repo — supported models list includes Z-Image Turbo and base — https://github.com/ostris/ai-toolkit
5. CitronLegacy config generator — full YAML with fp8, sigmoid timestep, adapter path, resolution buckets — https://github.com/citronlegacy/ai-toolkit-zimage-config-generator/blob/main/example_configs/zimage_AnubisCitronOC_ZIT_v3.yaml
6. Civitai: "Here's how to Quickly Train a Z-Image Turbo LoRA" — 9-image test, 3000 steps, 1 hr on 5090, adapter v1/v2, trigger word, sample settings — https://civitai.com/articles/23233/heres-how-to-quickly-train-a-z-image-turbo-lora
7. Civitai: "Z Image Turbo - Character LoRA Training Guide" — 20-image dataset, 19 neutral prompts, sigmoid timestep, cache embeddings, LR warning, 24 GB VRAM note — https://civitai.com/articles/23842/z-image-turbo-character-lora-training-guide-easy-how-to-guide
8. Civitai: "Z Image Turbo (ZIT) Anime Lora Training Experience" — tags vs captions, natural language prompting, Illustrious VRAM comparison, small-dataset failures — https://civitai.com/articles/24472/z-image-turbo-zit-anime-lora-training-experience
9. Civitai: "Z-Image Turbo LoRA training with AI Toolkit and Z-Image ControlNet" — ControlNet Union video chapters, FP8 scaled model 5.7 GB — https://civitai.com/articles/23314/z-image-turbo-lora-training-with-ai-toolkit-and-z-image-controlnet-full-tutorial-for-highest-quality
10. arXiv:2511.22699 — Z-Image paper, 314K H800 GPU hours training cost, Decoupled-DMD distillation — https://arxiv.org/abs/2511.22699
11. ComfyUI JoyCaption guide — local captioning models for NSFW/private datasets — https://www.runcomfy.com/comfyui-nodes/Comfyui_CXH_joy_caption

Further reading

• How to caption datasets for diffusion training
• AI Toolkit vs ComfyUI for LoRA training
• Understanding fp8 quantization in diffusion models