--- title: Ming-Image-0.1-Design-bf16 canonical_url: "https://www.modelscope.cn/models/mlx-community/Ming-Image-0.1-Design-bf16" md_url: "https://www.modelscope.cn/models/mlx-community/Ming-Image-0.1-Design-bf16.md" repository: mlx-community/Ming-Image-0.1-Design-bf16 last_updated: 2026-09-28 license: mit pipeline_tag: text-to-image-synthesis tasks: - text-to-image-synthesis base_model: - inclusionAI/Ming-Image-0.1-Design base_model_relation: finetune parameters: 24.9B tensor_type: - BF16 - F32 library_name: - mlx - lora - safetensors - diffusers - pytorch frameworks: - pytorch language: - en - zh downloads: 39 stars: 0 tags: - mlx - safetensors - apple-silicon - text-to-image - diffusion - s3-dit - rgba - graphic-design - ming-image --- # Ming-Image-0.1-Design-bf16 > Ming-Image-0.1-Design-bf16 - mlx-community 在 ModelScope 开源的模型。mlx-community/Ming-Image-0.1-Design-bf16 mlx-community/Ming-Image-0.1-Design-bf16 是 ModelScope 魔搭社区上的 24.9B 参数text-to-image-synthesis模型，采用 mit 许可，基于 inclusionAI/Ming-Image-0.1-Design 构建。 - **Repository**: mlx-community/Ming-Image-0.1-Design-bf16 - **License**: mit - **Tasks**: text-to-image-synthesis - **Parameters**: 24.9B - **Base model**: inclusionAI/Ming-Image-0.1-Design - **Tags**: mlx, safetensors, apple-silicon, text-to-image, diffusion, s3-dit, rgba, graphic-design, ming-image - **Downloads**: 39 - **Stars**: 0 - **Last updated**: 2026-09-28 Source: https://www.modelscope.cn/models/mlx-community/Ming-Image-0.1-Design-bf16 --- # mlx-community/Ming-Image-0.1-Design-bf16 MLX (bf16) snapshot of **[inclusionAI/Ming-Image-0.1-Design](https://huggingface.co/inclusionAI/Ming-Image-0.1-Design)** (MIT) for Apple Silicon, loaded by the Swift/MLX port [ming-image-swift](https://github.com/xocialize/ming-image-swift). Ming-Image-0.1-Design is a design-focused text-to-image model for posters, signage backings, cards, UI layouts and typography, and it generates **RGBA** images. A BailingMM2 MLLM (Ling-mini-2.0 MoE) conditions a Z-Image-style S3-DiT through learnable query tokens and a direct hidden-state stream; a 4-channel Qwen-Image VAE decodes colour and alpha. Sibling: [mlx-community/Ming-Image-0.1-Design-Layer-bf16](https://huggingface.co/mlx-community/Ming-Image-0.1-Design-Layer-bf16), which decomposes a flat design into editable RGBA layers. Quantized tiers of this model: - [8-bit](https://huggingface.co/mlx-community/Ming-Image-0.1-Design-8bit): 28.3 GB, matches bf16 on every quality gate, 48 GB Macs; - [4-bit](https://huggingface.co/mlx-community/Ming-Image-0.1-Design-4bit): 20.5 GB, a small measured cost, 36 GB Macs. ## What's in this repo The upstream diffusers-style tree, with each component stored at the precision the port runs it: | Component | Stored | vs upstream | |---|---|---| | `mllm/`: Ling-mini-2.0 MoE MLLM, Qwen2.5 ViT, tokenizer | bf16, 34.0 GB | byte-identical | | `connector/`: Qwen2-1.5B non-causal connector | bf16, 3.1 GB | cast from f32 | | `mlp/`: query tokens and projection heads | f32, 0.12 GB | byte-identical | | `transformer/`: S3-DiT | bf16, 12.3 GB | byte-identical | | `vae/`: RGBA VAE | bf16, 0.25 GB | byte-identical | Total: 49.8 GB. The one cast is MLX's own f32 → bf16 conversion, the same one the port applies when it loads the upstream checkpoint, so the port runs bit-identical parameters from either repo (checked tensor by tensor). ## Parity (Swift port vs the PyTorch reference) Components on the fp32 CPU parity lane, against reference goldens: | Component | Result | |---|---| | S3-DiT at 256² and 1024² | relL2 ≤ 1.5e-5 | | VAE decode / encode | relL2 9.1e-7 / 3.5e-6 | | Scheduler | bit-exact | | Tokenizer, chat template, 3-D position ids | exact | | MoE hidden states (with the reference's expert routing; ties make unforced routing implementation-defined) | ≤ 3.3e-6 | | Conditioning outputs (same routing) | ≤ 3.6e-5 | End to end: - 2-step fp32 generation: final latents relL2 3.5e-6, and the RGBA output is within 1 LSB. - Production bf16 on the GPU lands **41.9 dB** from the fp32-truth render at 1024². ## Native alpha Transparency is prompt-driven. The Swift package exposes it as a request field (`background: .transparent`, MLXEngine contract 1.48.0) and runs the recipe we measured on this port: - the phrase *"transparent canvas, not white, not checkerboard"* first; - a retry at the same seed with *"不要白底，不要棋盘格，只要透明通道"* when the alpha comes back opaque; - then an alpha floor (α < 0.1 → 0). That recipe gave usable alpha on 5 of the 6 test subjects. Glass and other transparent materials fail: they come back opaque, or with a checkerboard painted into the alpha. Matte those instead. ## Memory and speed (M5 Max) Measured as process `phys_footprint`, with MLX's buffer cache capped at 2 GB (MLXEngine's default). - **Post-load resident:** 13.2 GB (the DiT and the VAE). - **Peak:** 49.9 GB at 1024², 2048² and 2560×1440 alike. That peak is the conditioning stage: the 34 GB MLLM is loaded, conditions the prompt, and is released before denoising. The 2048²-class VAE decode is bounded, using a chunked mid-block attention and a tiled up path, and is exact. - **Under MLXEngine:** it declares 13.5 GB resident plus 44.1 GB activation. Against the engine's 0.7× memory budget, that needs a 96 GB Mac. On 48 GB use the 8-bit tier, and on 36 GB the 4-bit tier. - **Speed at 12 steps, including conditioning:** 39 s at 1024², 245 s at 2048². ## Use (Swift / MLXEngine) ```swift import MLXMingImage import MLXToolKit let package = MingImageT2IPackage(configuration: MingImageConfiguration(snapshotPath: "")) try await package.load() let response = try await package.run(T2IRequest( prompt: "A modern tech conference poster titled 'MLX SUMMIT 2026', bold geometric shapes", width: 1024, height: 1024, seed: 42)) as! T2IResponse // response.image: an RGBA PNG with straight alpha; pass background: .transparent for a cut-out asset ``` Under MLXEngine, register `MingImageT2IPackage.registration` with `MingImageConfiguration(repo: "mlx-community/Ming-Image-0.1-Design")`, and the engine downloads this repo on first use. ## License MIT, as the upstream weights. The upstream LICENSE is included.