Publish MOSS-Image SFT AnyFlow recipes, adapters, and validation previews

.gitattributes

@@ -713,3 +713,19 @@ validation_images/step_27500_portrait_0_512x512.png filter=lfs diff=lfs merge=lf
 validation_images/step_27500_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_27500_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_27500_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-28000/assets/step_28000_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_28000_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -12,7 +12,7 @@ tags:
 
 Forward MeanFlow pretraining. Stage 2 initializes from the final EMA adapter.
 
-Status: **training**. Observed updates: 27,600 / 100,000.
+Status: **training**. Observed updates: 28,130 / 100,000.
 
 ## Training recipe
 
@@ -49,11 +49,11 @@ Schedule correction: the stage-1 run resumed from update 1,000 with the [flow-sc
 
 [Pre-correction previews](pre-schedule-fix/validation_images) remain available.
 
-[1,000](checkpoint-1000), [2,000](checkpoint-2000), [3,000](checkpoint-3000), [4,000](checkpoint-4000), [5,000](checkpoint-5000), [6,000](checkpoint-6000), [7,000](checkpoint-7000), [8,000](checkpoint-8000), [9,000](checkpoint-9000), [10,000](checkpoint-10000), [11,000](checkpoint-11000), [12,000](checkpoint-12000), [13,000](checkpoint-13000), [14,000](checkpoint-14000), [15,000](checkpoint-15000), [16,000](checkpoint-16000), [17,000](checkpoint-17000), [18,000](checkpoint-18000), [19,000](checkpoint-19000), [20,000](checkpoint-20000), [21,000](checkpoint-21000), [22,000](checkpoint-22000), [23,000](checkpoint-23000), [24,000](checkpoint-24000), [25,000](checkpoint-25000), [26,000](checkpoint-26000), [27,000](checkpoint-27000)
+[1,000](checkpoint-1000), [2,000](checkpoint-2000), [3,000](checkpoint-3000), [4,000](checkpoint-4000), [5,000](checkpoint-5000), [6,000](checkpoint-6000), [7,000](checkpoint-7000), [8,000](checkpoint-8000), [9,000](checkpoint-9000), [10,000](checkpoint-10000), [11,000](checkpoint-11000), [12,000](checkpoint-12000), [13,000](checkpoint-13000), [14,000](checkpoint-14000), [15,000](checkpoint-15000), [16,000](checkpoint-16000), [17,000](checkpoint-17000), [18,000](checkpoint-18000), [19,000](checkpoint-19000), [20,000](checkpoint-20000), [21,000](checkpoint-21000), [22,000](checkpoint-22000), [23,000](checkpoint-23000), [24,000](checkpoint-24000), [25,000](checkpoint-25000), [26,000](checkpoint-26000), [27,000](checkpoint-27000), [28,000](checkpoint-28000)
 
 [All validation previews](validation_images) include live and EMA comparisons at both resolutions.
 
-Latest live adapter: [update 27,000](checkpoint-27000/pytorch_lora_weights.safetensors).
+Latest live adapter: [update 28,000](checkpoint-28000/pytorch_lora_weights.safetensors).
 
 ## Inference
 

checkpoint-28000/ema/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2275a7116a7823e3403276c0f02f593a2bcd4069a9a5d1e65c44b9661a4bd037
+size 294180952

checkpoint-28000/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1760d731e8e6426ba90feb151efb5451f626a4ef992486008b6dafa88ff068f7
+size 294180952

publication.json

@@ -107,10 +107,14 @@
     "27000": {
       "live": "ce4d205a148a7e1063c34a11862ca40e6ac050b2fba47b4ea637f75f8a8f9ca5",
       "ema": "ac16d7018872162e52da49f190f2caf1b1812f6bc259590a415c6491a47b14fb"
+    },
+    "28000": {
+      "live": "1760d731e8e6426ba90feb151efb5451f626a4ef992486008b6dafa88ff068f7",
+      "ema": "2275a7116a7823e3403276c0f02f593a2bcd4069a9a5d1e65c44b9661a4bd037"
     }
   },
   "state": "training",
-  "step": 27600,
+  "step": 28130,
   "previews": [
     "validation_images/step_0_character_0_512x512.png",
     "validation_images/step_0_character_1_1024x1024.png",
@@ -440,6 +444,14 @@
     "validation_images/step_27500_portrait_1_1024x1024.png",
     "validation_images/step_27500_text_0_512x512.png",
     "validation_images/step_27500_text_1_1024x1024.png",
+    "validation_images/step_28000_character_0_512x512.png",
+    "validation_images/step_28000_character_1_1024x1024.png",
+    "validation_images/step_28000_photo_0_512x512.png",
+    "validation_images/step_28000_photo_1_1024x1024.png",
+    "validation_images/step_28000_portrait_0_512x512.png",
+    "validation_images/step_28000_portrait_1_1024x1024.png",
+    "validation_images/step_28000_text_0_512x512.png",
+    "validation_images/step_28000_text_1_1024x1024.png",
     "validation_images/step_3000_character_0_512x512.png",
     "validation_images/step_3000_character_1_1024x1024.png",
     "validation_images/step_3000_photo_0_512x512.png",