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

.gitattributes

@@ -1241,3 +1241,19 @@ validation_images/step_49500_portrait_0_512x512.png filter=lfs diff=lfs merge=lf
 validation_images/step_49500_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_49500_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_49500_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-50000/assets/step_50000_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_50000_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: 49,630 / 100,000.
+Status: **training**. Observed updates: 50,010 / 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), [28,000](checkpoint-28000), [29,000](checkpoint-29000), [30,000](checkpoint-30000), [31,000](checkpoint-31000), [32,000](checkpoint-32000), [33,000](checkpoint-33000), [34,000](checkpoint-34000), [35,000](checkpoint-35000), [36,000](checkpoint-36000), [37,000](checkpoint-37000), [38,000](checkpoint-38000), [39,000](checkpoint-39000), [40,000](checkpoint-40000), [41,000](checkpoint-41000), [42,000](checkpoint-42000), [43,000](checkpoint-43000), [44,000](checkpoint-44000), [45,000](checkpoint-45000), [46,000](checkpoint-46000), [47,000](checkpoint-47000), [48,000](checkpoint-48000), [49,000](checkpoint-49000)
+[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), [29,000](checkpoint-29000), [30,000](checkpoint-30000), [31,000](checkpoint-31000), [32,000](checkpoint-32000), [33,000](checkpoint-33000), [34,000](checkpoint-34000), [35,000](checkpoint-35000), [36,000](checkpoint-36000), [37,000](checkpoint-37000), [38,000](checkpoint-38000), [39,000](checkpoint-39000), [40,000](checkpoint-40000), [41,000](checkpoint-41000), [42,000](checkpoint-42000), [43,000](checkpoint-43000), [44,000](checkpoint-44000), [45,000](checkpoint-45000), [46,000](checkpoint-46000), [47,000](checkpoint-47000), [48,000](checkpoint-48000), [49,000](checkpoint-49000), [50,000](checkpoint-50000)
 
 [All validation previews](validation_images) include live and EMA comparisons at both resolutions.
 
-Latest live adapter: [update 49,000](checkpoint-49000/pytorch_lora_weights.safetensors).
+Latest live adapter: [update 50,000](checkpoint-50000/pytorch_lora_weights.safetensors).
 
 ## Inference
 

checkpoint-50000/ema/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:483511c9c6b14cffc0941e0a49a570fae5bf2d7b160e0c9e27c1da1d7064956d
+size 294180952

checkpoint-50000/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0de208cc13f81b44c12844b008a586b3bed893ebe5a9aa1e6f4dd98d0a138922
+size 294180952

publication.json

@@ -195,10 +195,14 @@
     "49000": {
       "live": "b286cd60303343ecc2c71cb186fd079922011fe746266ae9a658d15f08991058",
       "ema": "6b3fc1b84e90801e052c172e14fa4abf71ec3afa94933b7196b5d9b0b188b8b6"
+    },
+    "50000": {
+      "live": "0de208cc13f81b44c12844b008a586b3bed893ebe5a9aa1e6f4dd98d0a138922",
+      "ema": "483511c9c6b14cffc0941e0a49a570fae5bf2d7b160e0c9e27c1da1d7064956d"
     }
   },
   "state": "training",
-  "step": 49630,
+  "step": 50010,
   "previews": [
     "validation_images/step_0_character_0_512x512.png",
     "validation_images/step_0_character_1_1024x1024.png",
@@ -912,6 +916,14 @@
     "validation_images/step_49500_portrait_1_1024x1024.png",
     "validation_images/step_49500_text_0_512x512.png",
     "validation_images/step_49500_text_1_1024x1024.png",
+    "validation_images/step_50000_character_0_512x512.png",
+    "validation_images/step_50000_character_1_1024x1024.png",
+    "validation_images/step_50000_photo_0_512x512.png",
+    "validation_images/step_50000_photo_1_1024x1024.png",
+    "validation_images/step_50000_portrait_0_512x512.png",
+    "validation_images/step_50000_portrait_1_1024x1024.png",
+    "validation_images/step_50000_text_0_512x512.png",
+    "validation_images/step_50000_text_1_1024x1024.png",
     "validation_images/step_5000_character_0_512x512.png",
     "validation_images/step_5000_character_1_1024x1024.png",
     "validation_images/step_5000_photo_0_512x512.png",