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

.gitattributes

@@ -329,3 +329,19 @@ validation_images/step_11500_portrait_0_512x512.png filter=lfs diff=lfs merge=lf
 validation_images/step_11500_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_11500_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
 validation_images/step_11500_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+checkpoint-12000/assets/step_12000_text_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_character_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_character_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_photo_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_photo_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_portrait_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_portrait_1_1024x1024.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_text_0_512x512.png filter=lfs diff=lfs merge=lfs -text
+validation_images/step_12000_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: 11,640 / 20,000.
+Status: **training**. Observed updates: 12,020 / 20,000.
 
 ## Training recipe
 
@@ -43,11 +43,11 @@ 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)
+[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)
 
 [All validation previews](validation_images) include live and EMA comparisons at both resolutions.
 
-Latest live adapter: [update 11,000](checkpoint-11000/pytorch_lora_weights.safetensors).
+Latest live adapter: [update 12,000](checkpoint-12000/pytorch_lora_weights.safetensors).
 
 ## Inference
 

checkpoint-12000/ema/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20b601a1bd78f9d4da29c4f4f143f3d9a01e49e1fbb76aed5d81b7abf3b8034b
+size 294180952

checkpoint-12000/pytorch_lora_weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c9cd4fbc7961f6bd77dc6139ad2bfa679595541d701414a1098ea8c760c33a
+size 294180952

publication.json

@@ -43,10 +43,14 @@
     "11000": {
       "live": "1ab37054d0d17fbfc5c805c88f46671aadfc4f349b236c62013d2c02b029d497",
       "ema": "c420451a1ab62c920a6b8c07351aa2b04d3213249e8791e4ed5edcbb0b80d458"
+    },
+    "12000": {
+      "live": "31c9cd4fbc7961f6bd77dc6139ad2bfa679595541d701414a1098ea8c760c33a",
+      "ema": "20b601a1bd78f9d4da29c4f4f143f3d9a01e49e1fbb76aed5d81b7abf3b8034b"
     }
   },
   "state": "training",
-  "step": 11640,
+  "step": 12020,
   "previews": [
     "validation_images/step_0_character_0_512x512.png",
     "validation_images/step_0_character_1_1024x1024.png",
@@ -96,6 +100,14 @@
     "validation_images/step_11500_portrait_1_1024x1024.png",
     "validation_images/step_11500_text_0_512x512.png",
     "validation_images/step_11500_text_1_1024x1024.png",
+    "validation_images/step_12000_character_0_512x512.png",
+    "validation_images/step_12000_character_1_1024x1024.png",
+    "validation_images/step_12000_photo_0_512x512.png",
+    "validation_images/step_12000_photo_1_1024x1024.png",
+    "validation_images/step_12000_portrait_0_512x512.png",
+    "validation_images/step_12000_portrait_1_1024x1024.png",
+    "validation_images/step_12000_text_0_512x512.png",
+    "validation_images/step_12000_text_1_1024x1024.png",
     "validation_images/step_1500_character_0_512x512.png",
     "validation_images/step_1500_character_1_1024x1024.png",
     "validation_images/step_1500_photo_0_512x512.png",