v2: core-compatible text encoder keys (model.*), core workflows, Comfy-Org comparison, VAE description fix

README.md

@@ -36,13 +36,15 @@ which is the better starting point for LoRA training. As a check, our converter
 | :-- | :-- | --: |
 | `diffusion_models/moss-image-base-dit-int8-convrot.safetensors` | DiT, INT8 ConvRot (W8, per-channel) | 5.41 GB (5,413,723,800 bytes) |
 | `diffusion_models/moss-image-base-dit-int4-convrot.safetensors` | DiT, INT4 ConvRot W4A4 | 3.04 GB (3,035,537,344 bytes) |
-| `text_encoders/moss-image-qwen35-int8-convrot.safetensors` | Qwen3.5 text encoder, INT8 ConvRot (optional) | 8.45 GB (8,452,951,644 bytes) |
-| `vae/moss-image-vae.safetensors` | FLUX.2 VAE, unchanged BF16 copy | 0.25 GB (249,521,372 bytes) |
-| `workflows/*.json` | ComfyUI workflows (UI and API format) for each DiT | |
+| `text_encoders/moss-image-qwen35-int8-convrot.safetensors` | Qwen3.5 text encoder, INT8 ConvRot (optional) | 8.45 GB (8,452,959,908 bytes) |
+| `vae/moss-image-vae.safetensors` | official MOSS-Image VAE (FLUX.2 latent space, slim decoder), unchanged FP32 copy | 0.25 GB (249,521,372 bytes) |
+| `workflows/core/*.json` | workflows for native ComfyUI core support (stock nodes), UI and API format | |
+| `workflows/*.json` | workflows for the MOSS-Image plugin, UI and API format | |
 | `quantize_moss.py` | conversion script | |
 
 The BF16 DiT is ~10.2 GB and the BF16 text encoder ~16.8 GB.
 MOSS-Image-Base and MOSS-Image-SFT ship byte-identical text encoder and VAE weights, so the text encoder here is identical to the one in the Abiray INT8 repo.
+The text encoder uses HF / ComfyUI-core key names (`model.language_model.*`, `model.visual.*`). Both ComfyUI core and the MOSS-Image plugin accept them.
 You can also keep the BF16 text encoder, exported with the plugin's `export.py`.
 
 ## Quantization method
@@ -102,12 +104,30 @@ which is FP32 in MOSS-Image-Base and is kept as FP32. The INT8 text encoder has
 - Both DiT files and the INT8 text encoder load through `MOSSImageUNETLoader` / `MOSSImageCLIPLoader`.
 - The files appear in the loader lists.
 - An API-format prompt (INT4 DiT + INT8 TE + VAE, low resolution, 2 steps) ran end to end and saved an image. The INT8 DiT generated an image the same way.
+- The same test passes on ComfyUI core with native MOSS-Image support (PR #17095 at `76d935d`) using only stock nodes.
+  The DiT layers load as native quantized tensors (`int8_tensorwise` / `convrot_w4a4`).
 
 On CPU, ComfyUI dequantizes instead of using the int8 / int4 GPU kernels. So this test confirms the files are valid and loadable,
 not GPU kernel behavior, output quality or speed.
 
 ## ComfyUI usage
 
+Two loaders work with these files. Use whichever matches your ComfyUI.
+
+### A. Native ComfyUI core (no custom node)
+
+Native MOSS-Image support is being added to ComfyUI core in [Comfy-Org/ComfyUI#17095](https://github.com/Comfy-Org/ComfyUI/pull/17095)
+(open at the time of writing). With a ComfyUI build that includes it:
+- **Load Diffusion Model** (`UNETLoader`): `moss-image-base-dit-int8-convrot.safetensors` or `...-int4-convrot.safetensors`, `weight_dtype = default`.
+- **Load CLIP** (`CLIPLoader`): `moss-image-qwen35-int8-convrot.safetensors`, **type = `moss_image`**.
+  You can also use Comfy-Org's `qwen3.5_9b_int8_convrot.safetensors`, which has numerically equivalent weights.
+- **Load VAE** (`VAELoader`): `moss-image-vae.safetensors`. Comfy-Org's `flux2-vae.safetensors` also works (see the comparison below).
+- Workflows: `workflows/core/moss-image-base-int8-convrot.json` / `...-int4-convrot.json`, plus `*-api.json` for the HTTP API.
+
+Core detects the separate q/k/v projections in these files automatically. Comfy-Org's own files use a fused `qkv`.
+
+### B. Official MOSS-Image plugin
+
 Requirements:
 - A recent ComfyUI with comfy-kitchen >= 0.2.37. It must have INT4 ConvRot (`convrot_w4a4`) and ConvRot INT8 support; tested with ComfyUI `bdc63e7`.
 - PyTorch built for CUDA 13.0 or newer, so the comfy-kitchen CUDA kernels are enabled.
@@ -124,7 +144,7 @@ On GPUs or devices without int8 compute, ComfyUI falls back to dequantize-then-m
              "vae/moss-image-vae.safetensors"]:
        hf_hub_download("SumomoLee/MOSS-Image-Base-ConvRot-ComfyUI", f, local_dir="/path/to/ComfyUI/models")
    ```
-2. Load `workflows/moss-image-base-int8-convrot.json` or `workflows/moss-image-base-int4-convrot.json` in ComfyUI. The `*-api.json` files are for the HTTP API.
+2. Plugin workflows: load `workflows/moss-image-base-int8-convrot.json` or `workflows/moss-image-base-int4-convrot.json` in ComfyUI. The `*-api.json` files are for the HTTP API.
 3. The nodes are **MOSS-Image Load Diffusion Model** (`MOSSImageUNETLoader`, keep `weight_dtype = default`), **MOSS-Image Load Qwen3.5**
    (`MOSSImageCLIPLoader`), and the standard **Load VAE**.
 
@@ -132,12 +152,44 @@ Recommended settings for the base model, from the official user guide: 50 steps,
 Euler / sgm_uniform, uncond skip-layer guidance on layer 22 (SkipLayerGuidanceDiTSimple), and 1 to 2 MP resolutions in multiples of 16.
 Recommended negative prompt: `nsfw, worst quality, bad anatomy, watermark, jpeg artifact`.
 
+## Comparison with Comfy-Org/MOSS-Image
+
+[Comfy-Org/MOSS-Image](https://huggingface.co/Comfy-Org/MOSS-Image) ships BF16 and INT8 ConvRot DiTs for Base and SFT, an INT8 ConvRot Qwen3.5-9B
+text encoder and `flux2-vae`. It does not ship INT4. Checked by reading tensors from both repos:
+
+- **INT8 DiT:** the same `int8_tensorwise` + ConvRot (group 256, per-channel FP32 scale) format.
+  - Comfy-Org fuses q/k/v into `proj.qkv`; these files keep them separate, which both core and the plugin accept.
+  - The quantized values are identical up to a handful of single-step rounding ties (fewer than 1 in 5 million values).
+  - Per-layer weight error is identical (for example 0.855% vs 0.855% on qkv).
+  - Kept-precision tensors are byte-identical, including FP32 `rope.bands`.
+- **Attention:** Comfy-Org's `*_int8_convrot_attn` files also carry `layers.N.comfy_attention.config = {"attention": "comfy_kitchen_int8"}`.
+  This opts each block into comfy-kitchen INT8 attention kernels where available. These files don't carry it, so attention runs at full precision:
+  slightly more accurate, possibly slower on GPUs with those kernels.
+- **Text encoder:** the same quantization and layer selection.
+  - Comfy-Org's file is the stock Qwen3.5-9B checkpoint. It adds `lm_head` and `mtp.*`, which conditioning doesn't use, and stores `A_log` and `linear_attn.norm` in FP32.
+  - This file is quantized from MOSS-Image's own text encoder, which stores those two in BF16 (the FP32 values rounded).
+  - The quantized weights agree to within rare one-step rounding ties.
+  - Core prints a harmless "Missing weight for layer model.lm_head" warning for this file; MOSS-Image never uses `lm_head`.
+- **VAE:** `moss-image-vae.safetensors` is the VAE shipped in MOSS-Image-Base. It is a FLUX.2 VAE with a slimmer decoder (96/192/384/384 channels, 62M params)
+  and slightly different encoder / batch-norm statistics from the stock `flux2-vae` (512-channel decoder, 84M params).
+  - The latent spaces are practically the same: under 0.4% difference in BN statistics and about 0.4% relative difference in encoded latents.
+  - The stock decoder reconstructs about 0.2–0.5 dB PSNR better in a quick test; the official decoder is smaller and faster.
+  - Either works.
+
+## Changelog
+
+- **v2:** the text encoder is re-keyed to `model.language_model.*` / `model.visual.*`. Tensor data is unchanged.
+  - The v1 key names (`language_model.*`, the plugin `export.py` naming) only load through the MOSS-Image plugin.
+  - In native ComfyUI core, v1 was silently misdetected as a CLIP-L model and produced invalid conditioning.
+  - v2 loads in both. Also: added core workflows, corrected the VAE description (FP32, not BF16), and added the Comfy-Org comparison. The DiT files are unchanged.
+- **v1:** initial release.
+
 ## Checksums (SHA-256)
 
 ```text
 1f42fac88a1b94f23ea3b49db6149243493ff1d6b962bb599a20417fe4aa71ea  diffusion_models/moss-image-base-dit-int4-convrot.safetensors
 f47c21a1397f846fa0d01c75e4b88fa6088783b47171399a465d92b1769ba0eb  diffusion_models/moss-image-base-dit-int8-convrot.safetensors
-41e7637a983b624f18c715f558fe24669c5d12d960331b655aeed50e7c609d6c  text_encoders/moss-image-qwen35-int8-convrot.safetensors
+511379fd690d59f11f49c062a8a687b9a985e92faa48a098eb7c66ba505574fe  text_encoders/moss-image-qwen35-int8-convrot.safetensors
 c3bdf94dc67ec69b30ea9fe1fa8340085679cae64d019f7c7e4a9921f1a7f17a  vae/moss-image-vae.safetensors
 ```
 

SHA256SUMS.txt

@@ -1,4 +1,4 @@
 1f42fac88a1b94f23ea3b49db6149243493ff1d6b962bb599a20417fe4aa71ea  diffusion_models/moss-image-base-dit-int4-convrot.safetensors
 f47c21a1397f846fa0d01c75e4b88fa6088783b47171399a465d92b1769ba0eb  diffusion_models/moss-image-base-dit-int8-convrot.safetensors
-41e7637a983b624f18c715f558fe24669c5d12d960331b655aeed50e7c609d6c  text_encoders/moss-image-qwen35-int8-convrot.safetensors
+511379fd690d59f11f49c062a8a687b9a985e92faa48a098eb7c66ba505574fe  text_encoders/moss-image-qwen35-int8-convrot.safetensors
 c3bdf94dc67ec69b30ea9fe1fa8340085679cae64d019f7c7e4a9921f1a7f17a  vae/moss-image-vae.safetensors

quantize_moss.py

@@ -72,6 +72,9 @@ def main():
                 cos = torch.nn.functional.cosine_similarity(deq.flatten(), wf.flatten(), dim=0).item()
                 stats.append((name, list(wf.shape), rel, cos))
                 print(f"{name} {list(wf.shape)} rel={rel:.5f} cos={cos:.6f}", flush=True)
+    if a.component == "te":
+        # HF / ComfyUI-core naming (model.language_model.*, model.visual.*); also accepted by the MOSS-Image plugin
+        out = {("model." + k if k.startswith(("language_model.", "visual.")) else k): v for k, v in out.items()}
     meta = {"format": "pt", "model": "moss-image"}
     if a.mode == "int4":
         meta["_quantization_metadata"] = json.dumps({"format_version": "1.0", "layers": layers})

text_encoders/moss-image-qwen35-int8-convrot.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41e7637a983b624f18c715f558fe24669c5d12d960331b655aeed50e7c609d6c
-size 8452951644
+oid sha256:511379fd690d59f11f49c062a8a687b9a985e92faa48a098eb7c66ba505574fe
+size 8452959908

workflows/core/moss-image-base-int4-convrot-api.json

@@ -0,0 +1,124 @@
+{
+ "1": {
+  "class_type": "UNETLoader",
+  "inputs": {
+   "unet_name": "moss-image-base-dit-int4-convrot.safetensors",
+   "weight_dtype": "default"
+  }
+ },
+ "2": {
+  "class_type": "CLIPLoader",
+  "inputs": {
+   "clip_name": "moss-image-qwen35-int8-convrot.safetensors",
+   "type": "moss_image",
+   "device": "default"
+  }
+ },
+ "3": {
+  "class_type": "VAELoader",
+  "inputs": {
+   "vae_name": "moss-image-vae.safetensors"
+  }
+ },
+ "4": {
+  "class_type": "CLIPTextEncode",
+  "inputs": {
+   "text": "A red fox sitting in a snowy pine forest at dawn.",
+   "clip": [
+    "2",
+    0
+   ]
+  }
+ },
+ "5": {
+  "class_type": "CLIPTextEncode",
+  "inputs": {
+   "text": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
+   "clip": [
+    "2",
+    0
+   ]
+  }
+ },
+ "6": {
+  "class_type": "EmptyFlux2LatentImage",
+  "inputs": {
+   "width": 1024,
+   "height": 1024,
+   "batch_size": 1
+  }
+ },
+ "15": {
+  "class_type": "ModelSamplingSD3",
+  "inputs": {
+   "shift": 5,
+   "model": [
+    "1",
+    0
+   ]
+  }
+ },
+ "13": {
+  "class_type": "SkipLayerGuidanceDiTSimple",
+  "inputs": {
+   "double_layers": "",
+   "single_layers": "22",
+   "start_percent": 0,
+   "end_percent": 1,
+   "model": [
+    "15",
+    0
+   ]
+  }
+ },
+ "16": {
+  "class_type": "KSampler",
+  "inputs": {
+   "seed": 42,
+   "steps": 50,
+   "cfg": 3,
+   "sampler_name": "euler",
+   "scheduler": "sgm_uniform",
+   "denoise": 1,
+   "model": [
+    "13",
+    0
+   ],
+   "positive": [
+    "4",
+    0
+   ],
+   "negative": [
+    "5",
+    0
+   ],
+   "latent_image": [
+    "6",
+    0
+   ]
+  }
+ },
+ "17": {
+  "class_type": "VAEDecode",
+  "inputs": {
+   "samples": [
+    "16",
+    0
+   ],
+   "vae": [
+    "3",
+    0
+   ]
+  }
+ },
+ "18": {
+  "class_type": "SaveImage",
+  "inputs": {
+   "filename_prefix": "moss-image-base",
+   "images": [
+    "17",
+    0
+   ]
+  }
+ }
+}

workflows/core/moss-image-base-int4-convrot.json

@@ -0,0 +1,577 @@
+{
+ "id": "602a9b4a-8b28-4c64-a19e-fb35024af9c8",
+ "revision": 2,
+ "last_node_id": 22,
+ "last_link_id": 30,
+ "nodes": [
+  {
+   "id": 4,
+   "type": "CLIPTextEncode",
+   "pos": [
+    590.980723946239,
+    282.97746942714315
+   ],
+   "size": [
+    430,
+    240
+   ],
+   "flags": {},
+   "order": 5,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "clip",
+     "type": "CLIP",
+     "link": 28
+    }
+   ],
+   "outputs": [
+    {
+     "name": "CONDITIONING",
+     "type": "CONDITIONING",
+     "slot_index": 0,
+     "links": [
+      18
+     ]
+    }
+   ],
+   "title": "moss-image Positive",
+   "properties": {
+    "Node name for S&R": "CLIPTextEncode"
+   },
+   "widgets_values": [
+    "A wide view of a refrigerator beside a kitchen counter with a fruit bowl. A small sticky note on the refrigerator door is handwritten in Chinese:\n\"记得买牛奶和鸡蛋！\n牛奶两盒，鸡蛋一打。\n顺便带一袋面包。\""
+   ],
+   "widgets_values_named": {
+    "text": "A wide view of a refrigerator beside a kitchen counter with a fruit bowl. A small sticky note on the refrigerator door is handwritten in Chinese:\n\"记得买牛奶和鸡蛋！\n牛奶两盒，鸡蛋一打。\n顺便带一袋面包。\""
+   }
+  },
+  {
+   "id": 5,
+   "type": "CLIPTextEncode",
+   "pos": [
+    592.3876751767972,
+    586.4494144007078
+   ],
+   "size": [
+    430,
+    200
+   ],
+   "flags": {},
+   "order": 6,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "clip",
+     "type": "CLIP",
+     "link": 29
+    }
+   ],
+   "outputs": [
+    {
+     "name": "CONDITIONING",
+     "type": "CONDITIONING",
+     "slot_index": 0,
+     "links": [
+      19
+     ]
+    }
+   ],
+   "title": "moss-image Negative",
+   "properties": {
+    "Node name for S&R": "CLIPTextEncode"
+   },
+   "widgets_values": [
+    "nsfw, worst quality, bad anatomy, watermark, jpeg artifact"
+   ],
+   "widgets_values_named": {
+    "text": ""
+   }
+  },
+  {
+   "id": 6,
+   "type": "EmptyFlux2LatentImage",
+   "pos": [
+    1076.8709415117667,
+    40.14629806728207
+   ],
+   "size": [
+    300,
+    180
+   ],
+   "flags": {},
+   "order": 0,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "LATENT",
+     "type": "LATENT",
+     "slot_index": 0,
+     "links": [
+      20
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "EmptyFlux2LatentImage"
+   },
+   "widgets_values": [
+    1024,
+    1024,
+    1
+   ],
+   "widgets_values_named": {
+    "width": 1024,
+    "height": 1024,
+    "batch_size": 1
+   }
+  },
+  {
+   "id": 13,
+   "type": "SkipLayerGuidanceDiTSimple",
+   "pos": [
+    1085.9671994568102,
+    810.0057777710338
+   ],
+   "size": [
+    340,
+    260
+   ],
+   "flags": {},
+   "order": 7,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 16
+    }
+   ],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "slot_index": 0,
+     "links": [
+      17
+     ]
+    }
+   ],
+   "title": "Uncond skip (empty = off)",
+   "properties": {
+    "Node name for S&R": "SkipLayerGuidanceDiTSimple"
+   },
+   "widgets_values": [
+    "",
+    "22",
+    0,
+    1
+   ],
+   "widgets_values_named": {
+    "double_layers": "",
+    "single_layers": "22",
+    "start_percent": 0,
+    "end_percent": 1
+   }
+  },
+  {
+   "id": 15,
+   "type": "ModelSamplingSD3",
+   "pos": [
+    635.5243515640974,
+    833.754663701251
+   ],
+   "size": [
+    310,
+    90
+   ],
+   "flags": {},
+   "order": 4,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 27
+    }
+   ],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "slot_index": 0,
+     "links": [
+      16
+     ]
+    }
+   ],
+   "title": "Official Flow sigma shift",
+   "properties": {
+    "Node name for S&R": "ModelSamplingSD3"
+   },
+   "widgets_values": [
+    5
+   ],
+   "widgets_values_named": {
+    "shift": 5
+   }
+  },
+  {
+   "id": 16,
+   "type": "KSampler",
+   "pos": [
+    1462.4773114331942,
+    579.4775120615249
+   ],
+   "size": [
+    270,
+    474
+   ],
+   "flags": {},
+   "order": 8,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 17
+    },
+    {
+     "name": "positive",
+     "type": "CONDITIONING",
+     "link": 18
+    },
+    {
+     "name": "negative",
+     "type": "CONDITIONING",
+     "link": 19
+    },
+    {
+     "name": "latent_image",
+     "type": "LATENT",
+     "link": 20
+    }
+   ],
+   "outputs": [
+    {
+     "name": "LATENT",
+     "type": "LATENT",
+     "links": [
+      22
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "KSampler"
+   },
+   "widgets_values": [
+    42,
+    "fixed",
+    50,
+    3,
+    "euler",
+    "sgm_uniform",
+    1
+   ],
+   "widgets_values_named": {
+    "seed": 42,
+    "control_after_generate": "fixed",
+    "steps": 50,
+    "cfg": 3,
+    "sampler_name": "euler",
+    "scheduler": "sgm_uniform",
+    "denoise": 1
+   }
+  },
+  {
+   "id": 17,
+   "type": "VAEDecode",
+   "pos": [
+    1866.0458086602723,
+    593.1343463769351
+   ],
+   "size": [
+    300,
+    180
+   ],
+   "flags": {},
+   "order": 9,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "samples",
+     "type": "LATENT",
+     "link": 22
+    },
+    {
+     "name": "vae",
+     "type": "VAE",
+     "link": 30
+    }
+   ],
+   "outputs": [
+    {
+     "name": "IMAGE",
+     "type": "IMAGE",
+     "slot_index": 0,
+     "links": [
+      21
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "VAEDecode"
+   }
+  },
+  {
+   "id": 18,
+   "type": "SaveImage",
+   "pos": [
+    2267.030358480412,
+    597.7554837229761
+   ],
+   "size": [
+    480,
+    600
+   ],
+   "flags": {},
+   "order": 10,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "images",
+     "type": "IMAGE",
+     "link": 21
+    }
+   ],
+   "outputs": [
+    {
+     "name": "images",
+     "type": "IMAGE",
+     "links": []
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "SaveImage"
+   },
+   "widgets_values": [
+    "moss-image"
+   ],
+   "widgets_values_named": {
+    "filename_prefix": "moss-image"
+   }
+  },
+  {
+   "id": 20,
+   "type": "UNETLoader",
+   "pos": [
+    108.25581677741914,
+    534.5600783323451
+   ],
+   "size": [
+    333.183203125,
+    82
+   ],
+   "flags": {},
+   "order": 1,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "links": [
+      27
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "UNETLoader"
+   },
+   "widgets_values": [
+    "moss-image-base-dit-int4-convrot.safetensors",
+    "default"
+   ],
+   "widgets_values_named": {
+    "unet_name": "moss-image-sft-dit-int8.safetensors",
+    "weight_dtype": "default"
+   }
+  },
+  {
+   "id": 21,
+   "type": "CLIPLoader",
+   "pos": [
+    103.86961675746264,
+    699.8493345306907
+   ],
+   "size": [
+    287.8609375,
+    82
+   ],
+   "flags": {},
+   "order": 2,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "CLIP",
+     "type": "CLIP",
+     "links": [
+      28,
+      29
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "CLIPLoader"
+   },
+   "widgets_values": [
+    "moss-image-qwen35-int8-convrot.safetensors",
+    "moss_image",
+    "default"
+   ],
+   "widgets_values_named": {
+    "clip_name": "moss-image-sft-qwen35-int8.safetensors",
+    "device": "default"
+   }
+  },
+  {
+   "id": 22,
+   "type": "VAELoader",
+   "pos": [
+    127.82638152549406,
+    874.154368339856
+   ],
+   "size": [
+    270,
+    58
+   ],
+   "flags": {},
+   "order": 3,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "VAE",
+     "type": "VAE",
+     "links": [
+      30
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "VAELoader"
+   },
+   "widgets_values": [
+    "moss-image-vae.safetensors"
+   ],
+   "widgets_values_named": {
+    "vae_name": "diffusion_pytorch_model.safetensors"
+   }
+  }
+ ],
+ "links": [
+  [
+   16,
+   15,
+   0,
+   13,
+   0,
+   "MODEL"
+  ],
+  [
+   17,
+   13,
+   0,
+   16,
+   0,
+   "MODEL"
+  ],
+  [
+   18,
+   4,
+   0,
+   16,
+   1,
+   "CONDITIONING"
+  ],
+  [
+   19,
+   5,
+   0,
+   16,
+   2,
+   "CONDITIONING"
+  ],
+  [
+   20,
+   6,
+   0,
+   16,
+   3,
+   "LATENT"
+  ],
+  [
+   21,
+   17,
+   0,
+   18,
+   0,
+   "IMAGE"
+  ],
+  [
+   22,
+   16,
+   0,
+   17,
+   0,
+   "LATENT"
+  ],
+  [
+   27,
+   20,
+   0,
+   15,
+   0,
+   "MODEL"
+  ],
+  [
+   28,
+   21,
+   0,
+   4,
+   0,
+   "CLIP"
+  ],
+  [
+   29,
+   21,
+   0,
+   5,
+   0,
+   "CLIP"
+  ],
+  [
+   30,
+   22,
+   0,
+   17,
+   1,
+   "VAE"
+  ]
+ ],
+ "groups": [],
+ "config": {},
+ "extra": {
+  "ds": {
+   "scale": 0.8052550000000055,
+   "offset": [
+    -622.1472402618422,
+    -410.32581838321397
+   ]
+  },
+  "frontendVersion": "1.55.17"
+ },
+ "version": 0.4
+}

workflows/core/moss-image-base-int8-convrot-api.json

@@ -0,0 +1,124 @@
+{
+ "1": {
+  "class_type": "UNETLoader",
+  "inputs": {
+   "unet_name": "moss-image-base-dit-int8-convrot.safetensors",
+   "weight_dtype": "default"
+  }
+ },
+ "2": {
+  "class_type": "CLIPLoader",
+  "inputs": {
+   "clip_name": "moss-image-qwen35-int8-convrot.safetensors",
+   "type": "moss_image",
+   "device": "default"
+  }
+ },
+ "3": {
+  "class_type": "VAELoader",
+  "inputs": {
+   "vae_name": "moss-image-vae.safetensors"
+  }
+ },
+ "4": {
+  "class_type": "CLIPTextEncode",
+  "inputs": {
+   "text": "A red fox sitting in a snowy pine forest at dawn.",
+   "clip": [
+    "2",
+    0
+   ]
+  }
+ },
+ "5": {
+  "class_type": "CLIPTextEncode",
+  "inputs": {
+   "text": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
+   "clip": [
+    "2",
+    0
+   ]
+  }
+ },
+ "6": {
+  "class_type": "EmptyFlux2LatentImage",
+  "inputs": {
+   "width": 1024,
+   "height": 1024,
+   "batch_size": 1
+  }
+ },
+ "15": {
+  "class_type": "ModelSamplingSD3",
+  "inputs": {
+   "shift": 5,
+   "model": [
+    "1",
+    0
+   ]
+  }
+ },
+ "13": {
+  "class_type": "SkipLayerGuidanceDiTSimple",
+  "inputs": {
+   "double_layers": "",
+   "single_layers": "22",
+   "start_percent": 0,
+   "end_percent": 1,
+   "model": [
+    "15",
+    0
+   ]
+  }
+ },
+ "16": {
+  "class_type": "KSampler",
+  "inputs": {
+   "seed": 42,
+   "steps": 50,
+   "cfg": 3,
+   "sampler_name": "euler",
+   "scheduler": "sgm_uniform",
+   "denoise": 1,
+   "model": [
+    "13",
+    0
+   ],
+   "positive": [
+    "4",
+    0
+   ],
+   "negative": [
+    "5",
+    0
+   ],
+   "latent_image": [
+    "6",
+    0
+   ]
+  }
+ },
+ "17": {
+  "class_type": "VAEDecode",
+  "inputs": {
+   "samples": [
+    "16",
+    0
+   ],
+   "vae": [
+    "3",
+    0
+   ]
+  }
+ },
+ "18": {
+  "class_type": "SaveImage",
+  "inputs": {
+   "filename_prefix": "moss-image-base",
+   "images": [
+    "17",
+    0
+   ]
+  }
+ }
+}

workflows/core/moss-image-base-int8-convrot.json

@@ -0,0 +1,577 @@
+{
+ "id": "602a9b4a-8b28-4c64-a19e-fb35024af9c8",
+ "revision": 2,
+ "last_node_id": 22,
+ "last_link_id": 30,
+ "nodes": [
+  {
+   "id": 4,
+   "type": "CLIPTextEncode",
+   "pos": [
+    590.980723946239,
+    282.97746942714315
+   ],
+   "size": [
+    430,
+    240
+   ],
+   "flags": {},
+   "order": 5,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "clip",
+     "type": "CLIP",
+     "link": 28
+    }
+   ],
+   "outputs": [
+    {
+     "name": "CONDITIONING",
+     "type": "CONDITIONING",
+     "slot_index": 0,
+     "links": [
+      18
+     ]
+    }
+   ],
+   "title": "moss-image Positive",
+   "properties": {
+    "Node name for S&R": "CLIPTextEncode"
+   },
+   "widgets_values": [
+    "A wide view of a refrigerator beside a kitchen counter with a fruit bowl. A small sticky note on the refrigerator door is handwritten in Chinese:\n\"记得买牛奶和鸡蛋！\n牛奶两盒，鸡蛋一打。\n顺便带一袋面包。\""
+   ],
+   "widgets_values_named": {
+    "text": "A wide view of a refrigerator beside a kitchen counter with a fruit bowl. A small sticky note on the refrigerator door is handwritten in Chinese:\n\"记得买牛奶和鸡蛋！\n牛奶两盒，鸡蛋一打。\n顺便带一袋面包。\""
+   }
+  },
+  {
+   "id": 5,
+   "type": "CLIPTextEncode",
+   "pos": [
+    592.3876751767972,
+    586.4494144007078
+   ],
+   "size": [
+    430,
+    200
+   ],
+   "flags": {},
+   "order": 6,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "clip",
+     "type": "CLIP",
+     "link": 29
+    }
+   ],
+   "outputs": [
+    {
+     "name": "CONDITIONING",
+     "type": "CONDITIONING",
+     "slot_index": 0,
+     "links": [
+      19
+     ]
+    }
+   ],
+   "title": "moss-image Negative",
+   "properties": {
+    "Node name for S&R": "CLIPTextEncode"
+   },
+   "widgets_values": [
+    "nsfw, worst quality, bad anatomy, watermark, jpeg artifact"
+   ],
+   "widgets_values_named": {
+    "text": ""
+   }
+  },
+  {
+   "id": 6,
+   "type": "EmptyFlux2LatentImage",
+   "pos": [
+    1076.8709415117667,
+    40.14629806728207
+   ],
+   "size": [
+    300,
+    180
+   ],
+   "flags": {},
+   "order": 0,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "LATENT",
+     "type": "LATENT",
+     "slot_index": 0,
+     "links": [
+      20
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "EmptyFlux2LatentImage"
+   },
+   "widgets_values": [
+    1024,
+    1024,
+    1
+   ],
+   "widgets_values_named": {
+    "width": 1024,
+    "height": 1024,
+    "batch_size": 1
+   }
+  },
+  {
+   "id": 13,
+   "type": "SkipLayerGuidanceDiTSimple",
+   "pos": [
+    1085.9671994568102,
+    810.0057777710338
+   ],
+   "size": [
+    340,
+    260
+   ],
+   "flags": {},
+   "order": 7,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 16
+    }
+   ],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "slot_index": 0,
+     "links": [
+      17
+     ]
+    }
+   ],
+   "title": "Uncond skip (empty = off)",
+   "properties": {
+    "Node name for S&R": "SkipLayerGuidanceDiTSimple"
+   },
+   "widgets_values": [
+    "",
+    "22",
+    0,
+    1
+   ],
+   "widgets_values_named": {
+    "double_layers": "",
+    "single_layers": "22",
+    "start_percent": 0,
+    "end_percent": 1
+   }
+  },
+  {
+   "id": 15,
+   "type": "ModelSamplingSD3",
+   "pos": [
+    635.5243515640974,
+    833.754663701251
+   ],
+   "size": [
+    310,
+    90
+   ],
+   "flags": {},
+   "order": 4,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 27
+    }
+   ],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "slot_index": 0,
+     "links": [
+      16
+     ]
+    }
+   ],
+   "title": "Official Flow sigma shift",
+   "properties": {
+    "Node name for S&R": "ModelSamplingSD3"
+   },
+   "widgets_values": [
+    5
+   ],
+   "widgets_values_named": {
+    "shift": 5
+   }
+  },
+  {
+   "id": 16,
+   "type": "KSampler",
+   "pos": [
+    1462.4773114331942,
+    579.4775120615249
+   ],
+   "size": [
+    270,
+    474
+   ],
+   "flags": {},
+   "order": 8,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "model",
+     "type": "MODEL",
+     "link": 17
+    },
+    {
+     "name": "positive",
+     "type": "CONDITIONING",
+     "link": 18
+    },
+    {
+     "name": "negative",
+     "type": "CONDITIONING",
+     "link": 19
+    },
+    {
+     "name": "latent_image",
+     "type": "LATENT",
+     "link": 20
+    }
+   ],
+   "outputs": [
+    {
+     "name": "LATENT",
+     "type": "LATENT",
+     "links": [
+      22
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "KSampler"
+   },
+   "widgets_values": [
+    42,
+    "fixed",
+    50,
+    3,
+    "euler",
+    "sgm_uniform",
+    1
+   ],
+   "widgets_values_named": {
+    "seed": 42,
+    "control_after_generate": "fixed",
+    "steps": 50,
+    "cfg": 3,
+    "sampler_name": "euler",
+    "scheduler": "sgm_uniform",
+    "denoise": 1
+   }
+  },
+  {
+   "id": 17,
+   "type": "VAEDecode",
+   "pos": [
+    1866.0458086602723,
+    593.1343463769351
+   ],
+   "size": [
+    300,
+    180
+   ],
+   "flags": {},
+   "order": 9,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "samples",
+     "type": "LATENT",
+     "link": 22
+    },
+    {
+     "name": "vae",
+     "type": "VAE",
+     "link": 30
+    }
+   ],
+   "outputs": [
+    {
+     "name": "IMAGE",
+     "type": "IMAGE",
+     "slot_index": 0,
+     "links": [
+      21
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "VAEDecode"
+   }
+  },
+  {
+   "id": 18,
+   "type": "SaveImage",
+   "pos": [
+    2267.030358480412,
+    597.7554837229761
+   ],
+   "size": [
+    480,
+    600
+   ],
+   "flags": {},
+   "order": 10,
+   "mode": 0,
+   "inputs": [
+    {
+     "name": "images",
+     "type": "IMAGE",
+     "link": 21
+    }
+   ],
+   "outputs": [
+    {
+     "name": "images",
+     "type": "IMAGE",
+     "links": []
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "SaveImage"
+   },
+   "widgets_values": [
+    "moss-image"
+   ],
+   "widgets_values_named": {
+    "filename_prefix": "moss-image"
+   }
+  },
+  {
+   "id": 20,
+   "type": "UNETLoader",
+   "pos": [
+    108.25581677741914,
+    534.5600783323451
+   ],
+   "size": [
+    333.183203125,
+    82
+   ],
+   "flags": {},
+   "order": 1,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "MODEL",
+     "type": "MODEL",
+     "links": [
+      27
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "UNETLoader"
+   },
+   "widgets_values": [
+    "moss-image-base-dit-int8-convrot.safetensors",
+    "default"
+   ],
+   "widgets_values_named": {
+    "unet_name": "moss-image-sft-dit-int8.safetensors",
+    "weight_dtype": "default"
+   }
+  },
+  {
+   "id": 21,
+   "type": "CLIPLoader",
+   "pos": [
+    103.86961675746264,
+    699.8493345306907
+   ],
+   "size": [
+    287.8609375,
+    82
+   ],
+   "flags": {},
+   "order": 2,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "CLIP",
+     "type": "CLIP",
+     "links": [
+      28,
+      29
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "CLIPLoader"
+   },
+   "widgets_values": [
+    "moss-image-qwen35-int8-convrot.safetensors",
+    "moss_image",
+    "default"
+   ],
+   "widgets_values_named": {
+    "clip_name": "moss-image-sft-qwen35-int8.safetensors",
+    "device": "default"
+   }
+  },
+  {
+   "id": 22,
+   "type": "VAELoader",
+   "pos": [
+    127.82638152549406,
+    874.154368339856
+   ],
+   "size": [
+    270,
+    58
+   ],
+   "flags": {},
+   "order": 3,
+   "mode": 0,
+   "inputs": [],
+   "outputs": [
+    {
+     "name": "VAE",
+     "type": "VAE",
+     "links": [
+      30
+     ]
+    }
+   ],
+   "properties": {
+    "Node name for S&R": "VAELoader"
+   },
+   "widgets_values": [
+    "moss-image-vae.safetensors"
+   ],
+   "widgets_values_named": {
+    "vae_name": "diffusion_pytorch_model.safetensors"
+   }
+  }
+ ],
+ "links": [
+  [
+   16,
+   15,
+   0,
+   13,
+   0,
+   "MODEL"
+  ],
+  [
+   17,
+   13,
+   0,
+   16,
+   0,
+   "MODEL"
+  ],
+  [
+   18,
+   4,
+   0,
+   16,
+   1,
+   "CONDITIONING"
+  ],
+  [
+   19,
+   5,
+   0,
+   16,
+   2,
+   "CONDITIONING"
+  ],
+  [
+   20,
+   6,
+   0,
+   16,
+   3,
+   "LATENT"
+  ],
+  [
+   21,
+   17,
+   0,
+   18,
+   0,
+   "IMAGE"
+  ],
+  [
+   22,
+   16,
+   0,
+   17,
+   0,
+   "LATENT"
+  ],
+  [
+   27,
+   20,
+   0,
+   15,
+   0,
+   "MODEL"
+  ],
+  [
+   28,
+   21,
+   0,
+   4,
+   0,
+   "CLIP"
+  ],
+  [
+   29,
+   21,
+   0,
+   5,
+   0,
+   "CLIP"
+  ],
+  [
+   30,
+   22,
+   0,
+   17,
+   1,
+   "VAE"
+  ]
+ ],
+ "groups": [],
+ "config": {},
+ "extra": {
+  "ds": {
+   "scale": 0.8052550000000055,
+   "offset": [
+    -622.1472402618422,
+    -410.32581838321397
+   ]
+  },
+  "frontendVersion": "1.55.17"
+ },
+ "version": 0.4
+}