LLaDA-Image: InclusionAI Open 6B Image Generation and Editing

ComfyUI Wikinews

InclusionAI releases LLaDA-Image, a 6B unified generation and editing model with a 4-step Turbo variant, plus a community ComfyUI node pack with INT8 and GGUF weights.

LLaDA-Image (GitHub, Hugging Face) is a 6B-parameter unified image generation and editing model family from InclusionAI (the Ant Group team behind Ling), released on September 4 under Apache-2.0 with a full open training recipe. A community ComfyUI node pack from RebelAI already runs it locally.
LLaDA-Image photorealistic generation showcase

Photorealistic generation samples from the official repository: natural lighting, lifelike detail and coherent multi-subject scenes.

One checkpoint for generating and editing

LLaDA-Image is a unified diffusion model where both the backbone and the DiT are diffusion models trained in a single framework. Unlike editor add-ons bolted onto a text-to-image model, a single checkpoint handles:

  • Text-to-image generation in 50 steps on the Base model
  • Instruction-guided editing with reference preservation, using the model's native editing path rather than a denoise-strength img2img workaround
  • VQ-conditioned generation through the SigVQ conditioning module
  • Chinese and English text rendering in generated images

The training recipe is fully open in the accompanying arXiv report: image-only pre-training builds the visual prior first, paired language supervision comes next, and joint generation-editing training unifies the two skills. On Qwen-Image-Bench it scores 53.53 in English and 53.38 in Chinese, state-of-the-art overall results for an open 6B model at release.

Text rendering and poster generation showcase

Chinese and English text rendering with poster-grade layouts.

Instruction-guided editing showcase

Native editing: content stays faithful while only the instructed change is applied.

Turbo: 4 steps instead of 50

Alongside the Base model, LLaDA-Image-Turbo distills the pipeline with Twin-DMD distillation down to 2 to 4 sampling steps for both generation and editing. Four checkpoint variants ship on day one: BF16 and FP8 for each of Base and Turbo, about 49 GB per variant in diffusers layout.

ModelStepsCheckpoints
LLaDA-Image (Base)50BF16, FP8
LLaDA-Image-Turbo2-4BF16, FP8
Qwen-Image-Bench comparison

Qwen-Image-Bench overall scores: LLaDA-Image tops the open 6B class in both English and Chinese.

Run it in ComfyUI with the RebelAI node pack

A community integration from RealRebelAI landed the same day the weights dropped. The pack provides four nodes: LLaDA Image Loader, LLaDA Image Text to Image, LLaDA Image Edit and LLaDA Image Unload. The edit node calls the model's native editing path (generation_mode="editing"), not a denoise-strength approximation, and edit dimensions must be divisible by 32.

The integration uses optimized weights from RealRebelAI/LLaDa-Image-Turbo_ComfyUI and a matching Base pack, about 29 GB per pack:

FilePlacementSize
LLaDA-Image-Turbo-transformer-BF16.safetensorsmodels/diffusion_models/13.1 GB
LLaDA-Image-Turbo-INT8.safetensorsmodels/diffusion_models/6.6 GB
LLaDA-Image-Turbo-text_encoder-Q4_K_M.ggufmodels/text_encoders/9.2 GB
LLaDa_VAE.safetensorsmodels/vae/0.17 GB

The INT8 transformer is a native safetensors quantization rather than GGUF; GGUF is used only for the quantized LLaDA2-MoE text encoder. For Turbo, start with 4 steps and CFG 1.0.

Training code is listed as coming soon on the official roadmap.

Comments

Sign in with GitHub to join the discussion.

Loading comments…
LLaDA-Image: InclusionAI Open 6B Image Generation and Editing | ComfyUI Wiki