Upload Z-Image-Turbo Fuliji LoRA adapter (rank=32, 3000 steps, 8 artists)

README.md

@@ -0,0 +1,240 @@
+---
+base_model: Tongyi-MAI/Z-Image-Turbo
+library_name: peft
+tags:
+  - lora
+  - diffusers
+  - text-to-image
+  - anime
+  - art-style
+  - z-image
+  - fuliji
+license: apache-2.0
+language:
+  - zh
+  - en
+---
+
+# Z-Image-Turbo × Fuliji — LoRA Adapter
+
+A **PEFT LoRA adapter** trained on top of [Tongyi-MAI/Z-Image-Turbo](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo) to learn the visual identity of 8 Chinese anime/illustration artists from the Fuliji dataset.
+
+> **Looking for the ready-to-run merged model?**
+> Use [DownFlow/Z-Image-Turbo-Fuli](https://huggingface.co/DownFlow/Z-Image-Turbo-Fuli) — the LoRA weights have been baked into the base model and can be served directly.
+
+---
+
+## Adapter Details
+
+| Property | Value |
+|---|---|
+| Base model | `Tongyi-MAI/Z-Image-Turbo` |
+| LoRA rank | 32 |
+| LoRA alpha | 32 |
+| Target modules | `to_q`, `to_k`, `to_v`, `w1`, `w2`, `w3` |
+| Trainable params | ~39 M |
+| Adapter size | ~271 MB |
+| Training steps | 3 000 |
+| Training resolution | 512 × 512 |
+| Dataset | [DownFlow/fuliji](https://huggingface.co/datasets/DownFlow/fuliji) (8 artists, ~200 images) |
+
+---
+
+## Quick Start (Python + Diffusers + PEFT)
+
+### 1 — Install dependencies
+
+```bash
+pip install diffusers transformers peft accelerate safetensors
+```
+
+### 2 — Generate with artist trigger token
+
+```python
+import torch
+from diffusers import DiffusionPipeline
+from peft import PeftModel
+
+DEVICE = "cuda"
+BASE_MODEL = "Tongyi-MAI/Z-Image-Turbo"
+ADAPTER = "DownFlow/Z-Image-Turbo-Fuli-LoRA"
+
+# Load base pipeline
+pipe = DiffusionPipeline.from_pretrained(
+    BASE_MODEL,
+    torch_dtype=torch.bfloat16,
+).to(DEVICE)
+
+# Attach LoRA adapter to the transformer
+pipe.transformer = PeftModel.from_pretrained(pipe.transformer, ADAPTER)
+
+# Generate — prepend the artist's trigger token
+# Trained artists: 萌芽儿o0, 年年, 封疆疆v, 焖焖碳, 星之迟迟, 蠢沫沫, 雨波HaneAme, 清水由乃
+image = pipe(
+    prompt="by 蠢沫沫, 1girl, solo, smile, looking at viewer, soft lighting",
+    num_inference_steps=8,
+    guidance_scale=0.0,   # Z-Image Turbo uses CFG=0
+    height=512,
+    width=512,
+).images[0]
+
+image.save("output.png")
+```
+
+### 3 — Adjust LoRA influence at runtime
+
+PEFT exposes a scaling multiplier per adapter. Increase it to push the style harder:
+
+```python
+# After PeftModel.from_pretrained ...
+for module in pipe.transformer.modules():
+    if hasattr(module, "scaling"):
+        module.scaling = {k: v * 1.5 for k, v in module.scaling.items()}
+```
+
+Recommended range: **1.0 – 2.0**. Values above 3.0 may cause colour artefacts.
+
+---
+
+## Merge and Unload (for maximum inference speed)
+
+Baking the LoRA into the base weights eliminates PEFT overhead entirely:
+
+```python
+import torch
+from diffusers import DiffusionPipeline
+from peft import PeftModel
+
+pipe = DiffusionPipeline.from_pretrained(
+    "Tongyi-MAI/Z-Image-Turbo",
+    torch_dtype=torch.bfloat16,
+)
+
+pipe.transformer = PeftModel.from_pretrained(
+    pipe.transformer,
+    "DownFlow/Z-Image-Turbo-Fuli-LoRA",
+)
+pipe.transformer = pipe.transformer.merge_and_unload()
+
+pipe.to("cuda")
+
+image = pipe(
+    prompt="by 年年, 1girl, white dress, cherry blossoms",
+    num_inference_steps=8,
+    guidance_scale=0.0,
+).images[0]
+```
+
+---
+
+## Serving with vLLM
+
+vLLM (≥ 0.8) supports serving diffusion pipelines via an OpenAI-compatible `/v1/images/generations` endpoint.
+
+> **Recommended flow for vLLM**: use the pre-merged model so no PEFT dependency is needed at serve time.
+
+### Option A — Serve the merged model (recommended)
+
+```bash
+pip install "vllm>=0.8.0"
+
+vllm serve DownFlow/Z-Image-Turbo-Fuli \
+    --task generate \
+    --dtype bfloat16 \
+    --max-model-len 512 \
+    --port 8000
+```
+
+Then call the endpoint:
+
+```bash
+curl http://localhost:8000/v1/images/generations \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "DownFlow/Z-Image-Turbo-Fuli",
+    "prompt": "by 蠢沫沫, 1girl, smile, soft watercolour style",
+    "n": 1,
+    "size": "512x512"
+  }'
+```
+
+Or via the OpenAI Python SDK:
+
+```python
+from openai import OpenAI
+
+client = OpenAI(base_url="http://localhost:8000/v1", api_key="not-needed")
+
+response = client.images.generate(
+    model="DownFlow/Z-Image-Turbo-Fuli",
+    prompt="by 年年, 1girl, white dress, cherry blossoms",
+    n=1,
+    size="512x512",
+)
+print(response.data[0].url)
+```
+
+### Option B — Serve with dynamic LoRA (experimental)
+
+vLLM supports dynamic LoRA module loading for LLMs; diffusion pipeline LoRA support is still experimental. If your vLLM build supports `--enable-lora` for image models:
+
+```bash
+vllm serve Tongyi-MAI/Z-Image-Turbo \
+    --task generate \
+    --dtype bfloat16 \
+    --enable-lora \
+    --lora-modules "fuliji=DownFlow/Z-Image-Turbo-Fuli-LoRA" \
+    --port 8000
+```
+
+Request with the LoRA active:
+
+```bash
+curl http://localhost:8000/v1/images/generations \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "fuliji",
+    "prompt": "by 雨波HaneAme, 1girl, beach, summer",
+    "n": 1,
+    "size": "512x512"
+  }'
+```
+
+---
+
+## Trained Artist Trigger Tokens
+
+Prepend `by <artist>, ` at the start of your prompt.
+
+| Token | Approx. images in training set |
+|---|---|
+| `萌芽儿o0` | 30 |
+| `年年` | 26 |
+| `封疆疆v` | 26 |
+| `焖焖碳` | 26 |
+| `星之迟迟` | 25 |
+| `蠢沫沫` | 23 |
+| `雨波HaneAme` | 23 |
+| `清水由乃` | 21 |
+
+---
+
+## Training Details
+
+- **Base model**: `Tongyi-MAI/Z-Image-Turbo` (8-step flow matching, CFG-free)
+- **Method**: PEFT LoRA, rank=32, alpha=32, dropout=0.05
+- **Dataset**: `DownFlow/fuliji` filtered to artists with ≥ 21 images
+- **Steps**: 3 000 with EMA (decay=0.9999)
+- **Optimizer**: AdamW, lr=1e-4, warmup=100 steps
+- **Batch**: 1 × 4 gradient accumulation = effective batch 4
+- **Augmentation**: horizontal flip, caption dropout 5%, timestep bias 1.2
+- **Regularisation**: 25% of batches sample from a 277-image generic dataset
+- **Hardware**: AMD MI300X, ROCm 6.2, bf16
+
+---
+
+## Related
+
+- [DownFlow/Z-Image-Turbo-Fuli](https://huggingface.co/DownFlow/Z-Image-Turbo-Fuli) — merged model (LoRA baked in, ready for `vllm serve`)
+- [DownFlow/fuliji](https://huggingface.co/datasets/DownFlow/fuliji) — training dataset
+- [Tongyi-MAI/Z-Image-Turbo](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo) — base model

adapter_config.json

@@ -0,0 +1,48 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "ZImageTransformer2DModel",
+    "parent_library": "diffusers.models.transformers.transformer_z_image"
+  },
+  "base_model_name_or_path": "Tongyi-MAI/Z-Image-Turbo",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32.0,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "to_v",
+    "w2",
+    "w1",
+    "w3",
+    "to_q",
+    "to_k"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c8f4c3c5e11f4f0782d4d472a24bfb5954dc1744150c3eace4297f88b8e78d
+size 284151432