> ## Documentation Index > Fetch the complete documentation index at: https://docs.sglang.io/llms.txt > Use this file to discover all available pages before exploring further. # Z-Image-Turbo > Deploy Z-Image-Turbo with SGLang - community contribution guide for Z-Image's fast image generation model. ## 1. Quick start Follow the [SGLang Diffusion installation guide](https://docs.sglang.io/docs/sglang-diffusion/installation.md) for your GPU platform. ## 2. Model capabilities [Z-Image-Turbo](https://huggingface.co/Tongyi-MAI/Z-Image-Turbo) is a distilled 6B single-stream DiT for fast text-to-image generation. It reaches its intended operating point in 8 function evaluations and is particularly strong at photorealistic scenes, prompt adherence, and English/Chinese text rendering. Choose it when latency and a relatively small deployment footprint matter more than the editability or maximum capacity of larger image models. It is a generation-only checkpoint; use Qwen-Image-Edit or FLUX.2 when the request includes source images or identity-preserving edits. ## 3. Model Deployment This section provides deployment configurations optimized for different hardware platforms and use cases. ### 3.1 Basic Configuration Z-Image-Turbo is optimized for high-quality image generation with only 8 inference steps. The recommended launch configurations vary by hardware. Use the [command builder](#1-quick-start) above. ### 3.2 Configuration Tips See [Performance Optimization](https://docs.sglang.io/docs/sglang-diffusion/performance-optimization.md) for acceleration features and their runtime requirements. * `--vae-path`: Path to a custom VAE model or HuggingFace model ID (e.g., fal/FLUX.2-Tiny-AutoEncoder). If not specified, the VAE will be loaded from the main model path. * `--num-gpus`: Number of GPUs to use * `--tp-size`: Tensor parallelism size (only for the encoder; should not be larger than 1 if text encoder offload is enabled, as layer-wise offload plus prefetch is faster) * `--sp-degree`: Sequence parallelism size (typically should match the number of GPUs) * `--ulysses-degree`: The degree of DeepSpeed-Ulysses-style SP in USP * `--ring-degree`: The degree of ring attention-style SP in USP **AMD ROCm Notes**: Requires SGLang >= v0.5.8. ## 4. API Usage For complete API documentation, please refer to the [official API usage guide](https://docs.sglang.io/docs/sglang-diffusion/api/openai_api.md). ### 4.1 Generate an Image ```python Example theme={null} import base64 from openai import OpenAI client = OpenAI(api_key="EMPTY", base_url="http://localhost:30000/v1") response = client.images.generate( model="Tongyi-MAI/Z-Image-Turbo", prompt="A logo With Bold Large text: SGL Diffusion", n=1, response_format="b64_json", ) # Save the generated image image_bytes = base64.b64decode(response.data[0].b64_json) with open("output.png", "wb") as f: f.write(image_bytes) ``` ### 4.2 Advanced Usage Z-Image uses `positive + guidance_scale * (positive - negative)` guidance, unlike the standard CFG convention. Any positive `guidance_scale` enables CFG; omitted negative prompts use an empty string. Z-Image-Turbo defaults to `guidance_scale=0` and does not run the negative branch. Enabling CFG adds a second DiT branch, so expect additional inference cost; it is not automatically recommended for Turbo. #### 4.2.1 Cache-DiT Acceleration SGLang integrates [Cache-DiT](https://github.com/vipshop/cache-dit), a caching acceleration engine for Diffusion Transformers (DiT), to achieve up to 7.4x inference speedup with minimal quality loss. You can set `SGLANG_CACHE_DIT_ENABLED=True` to enable it. For more details, please refer to the SGLang Cache-DiT [documentation](https://docs.sglang.io/docs/sglang-diffusion/cache_dit.md). **Basic Usage** ```bash Command theme={null} SGLANG_CACHE_DIT_ENABLED=true sglang serve --model-path Tongyi-MAI/Z-Image-Turbo ``` **Advanced Usage** * DBCache Parameters: DBCache controls block-level caching behavior:
Parameter Env Variable Default Description
Fn `SGLANG_CACHE_DIT_FN` 1 Number of first blocks to always compute
Bn `SGLANG_CACHE_DIT_BN` 0 Number of last blocks to always compute
W `SGLANG_CACHE_DIT_WARMUP` 4 Warmup steps before caching starts
R `SGLANG_CACHE_DIT_RDT` 0.24 Residual difference threshold
MC `SGLANG_CACHE_DIT_MC` 3 Maximum continuous cached steps
* TaylorSeer Configuration: TaylorSeer improves caching accuracy using Taylor expansion:
Parameter Env Variable Default Description
Enable `SGLANG_CACHE_DIT_TAYLORSEER` false Enable TaylorSeer calibrator
Order `SGLANG_CACHE_DIT_TS_ORDER` 1 Taylor expansion order (1 or 2)
Combined Configuration Example: ```bash Command theme={null} SGLANG_CACHE_DIT_ENABLED=true \ SGLANG_CACHE_DIT_FN=2 \ SGLANG_CACHE_DIT_BN=1 \ SGLANG_CACHE_DIT_WARMUP=4 \ SGLANG_CACHE_DIT_RDT=0.4 \ SGLANG_CACHE_DIT_MC=4 \ SGLANG_CACHE_DIT_TAYLORSEER=true \ SGLANG_CACHE_DIT_TS_ORDER=2 \ sglang serve --model-path Tongyi-MAI/Z-Image-Turbo ``` #### 4.2.2 CPU Offload * `--dit-cpu-offload`: Use CPU offload for DiT inference. Enable if run out of memory. * `--text-encoder-cpu-offload`: Use CPU offload for text encoder inference. * `--vae-cpu-offload`: Use CPU offload for VAE. * `--pin-cpu-memory`: Pin memory for CPU offload. Only added as a temp workaround if it throws "CUDA error: invalid argument". #### 4.2.3 Known LoRA examples Use `--lora-path` at startup or the [LoRA management API](https://docs.sglang.io/docs/sglang-diffusion/api/openai_api.md#lora-management) to load an adapter. Known Z-Image-Turbo examples include: * [`tarn59/pixel_art_style_lora_z_image_turbo`](https://huggingface.co/tarn59/pixel_art_style_lora_z_image_turbo) * [`wcde/Z-Image-Turbo-DeJPEG-Lora`](https://huggingface.co/wcde/Z-Image-Turbo-DeJPEG-Lora) ## 5. Benchmark Test Environment: * Hardware: AMD Instinct MI300X GPU (1x) * Model: Tongyi-MAI/Z-Image-Turbo * Docker Image: lmsysorg/sglang:v0.5.8-rocm700-mi30x * sglang diffusion version: 0.5.8 ### 5.1 Speedup Benchmark #### 5.1.1 Generate an image **AMD MI300X** **Server Command**: ```shell Command theme={null} sglang serve --model-path Tongyi-MAI/Z-Image-Turbo \ --ulysses-degree=1 --ring-degree=1 --port 30000 ``` **Benchmark Command**: ```shell Command theme={null} python3 -m sglang.multimodal_gen.benchmarks.bench_serving \ --dataset vbench --task text-to-image --num-prompts 1 --max-concurrency 1 ``` **Result**: ```text Output theme={null} ================= Serving Benchmark Result ================= Task: text-to-image Model: Tongyi-MAI/Z-Image-Turbo Dataset: vbench -------------------------------------------------- Benchmark duration (s): 1.84 Request rate: inf Max request concurrency: 1 Successful requests: 1/1 -------------------------------------------------- Request throughput (req/s): 0.54 Latency Mean (s): 1.8435 Latency Median (s): 1.8435 Latency P99 (s): 1.8435 -------------------------------------------------- Peak Memory Max (MB): 30689.20 Peak Memory Mean (MB): 30689.20 Peak Memory Median (MB): 30689.20 ============================================================ ``` **Ascend A3 Series** **Server Command**: ```shell Command theme={null} #One A3 Series card has 2 npu chips sglang serve --model-path Tongyi-MAI/Z-Image-Turbo --tp-size 2 --sp-degree 1 --num-gpus 2 ``` **Benchmark Command**: ```shell Command theme={null} python -m sglang.multimodal_gen.benchmarks.bench_serving --dataset vbench --task text-to-image --num-prompts 1 --max-concurrency 1 ``` **Result**: ```text Output theme={null} ================= Serving Benchmark Result ================= Task: text-to-image Model: Tongyi-MAI/Z-Image-Turbo Dataset: vbench -------------------------------------------------- Benchmark duration (s): 2.43 Request rate: inf Max request concurrency: 1 Successful requests: 1/1 Completed outputs: 1 Outputs per prompt: 1 -------------------------------------------------- Request throughput (req/s): 0.41 Output throughput (outputs/s): 0.41 Latency Mean (s): 2.43 Latency Median (s): 2.43 Latency P90 (s): 2.43 Latency P95 (s): 2.43 Latency P99 (s): 2.43 -------------------------------------------------- Peak Memory Max (MB): 11052.00 Peak Memory Mean (MB): 11052.00 Peak Memory Median (MB): 11052.00 ------------------------------------------------------------ ``` #### 5.1.2 Generate images with high concurrency **AMD MI300X** **Benchmark Command**: ```shell Command theme={null} python3 -m sglang.multimodal_gen.benchmarks.bench_serving \ --dataset vbench --task text-to-image --num-prompts 20 --max-concurrency 20 ``` **Result**: ```text Output theme={null} ================= Serving Benchmark Result ================= Task: text-to-image Model: Tongyi-MAI/Z-Image-Turbo Dataset: vbench -------------------------------------------------- Benchmark duration (s): 35.32 Request rate: inf Max request concurrency: 20 Successful requests: 20/20 -------------------------------------------------- Request throughput (req/s): 0.57 Latency Mean (s): 18.5672 Latency Median (s): 18.5573 Latency P99 (s): 34.9880 -------------------------------------------------- Peak Memory Max (MB): 30689.26 Peak Memory Mean (MB): 30689.21 Peak Memory Median (MB): 30689.21 ============================================================ ``` **Ascend A3 Series** **Benchmark Command**: ```shell Command theme={null} python -m sglang.multimodal_gen.benchmarks.bench_serving --dataset vbench --task text-to-image --num-prompts 20 --max-concurrency 20 ``` **Result**: ```text Output theme={null} ================= Serving Benchmark Result ================= Task: text-to-image Model: /models/Tongyi-MAI/Z-Image-Turbo/Z-Image-Turbo Dataset: vbench -------------------------------------------------- Benchmark duration (s): 49.08 Request rate: inf Max request concurrency: 20 Successful requests: 20/20 Completed outputs: 20 Outputs per prompt: 1 -------------------------------------------------- Request throughput (req/s): 0.41 Output throughput (outputs/s): 0.41 Latency Mean (s): 25.78 Latency Median (s): 25.77 Latency P90 (s): 44.42 Latency P95 (s): 46.75 Latency P99 (s): 48.61 -------------------------------------------------- Peak Memory Max (MB): 11054.00 Peak Memory Mean (MB): 11054.00 Peak Memory Median (MB): 11054.00 ------------------------------------------------------------ ``` ## 6. Run in ComfyUI Run this model from ComfyUI with the [SGLDiffusion plugin](https://github.com/sgl-project/sglang/tree/main/python/sglang/multimodal_gen/apps/ComfyUI_SGLDiffusion), which ships in the SGLang repository at `python/sglang/multimodal_gen/apps/ComfyUI_SGLDiffusion`. **Server mode** — SGLang runs the pipeline and ComfyUI sends the request. Start a server as shown above, point the `SGLDiffusion Server Model` node at it, then generate with `SGLDiffusion Generate Image`. **Integrated mode** — ComfyUI's own sampler, CLIP, and VAE drive the loop while SGLang replaces the model forward. Load the checkpoint with `SGLDiffusion UNET Loader` and set `model_type` to `lumina2` on the `SGLDiffusion Options` node. A reference workflow is included at `python/sglang/multimodal_gen/apps/ComfyUI_SGLDiffusion/workflows/z-image_sgld.json`. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.