Upload 4 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Photanima[[:space:]]v2.1[[:space:]]Turbo[[:space:]]FP16[[:space:]]-[[:space:]]Anima.png filter=lfs diff=lfs merge=lfs -text

Photanima v2.0 FP16 - Anima - 6-8steps,er_sde sampler on ode mode.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b44b170a32b922c82c50bf35e52206e9003396ba4bfd1a295b5738351da97452
+size 4182711376

Photanima v2.1 Turbo FP16 - Anima.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6236ce18de79ca62b45ab4f0e0d5cfd0b06370acd0c16c2c98c07a59a196279
+size 4182268256

Photanima v2.1 Turbo FP16 - Anima.txt

@@ -0,0 +1,69 @@
+Photanima v2.1 Turbo FP16 - Anima
+
+ Technical details
+
+v2 is trained on ~2000 images for 45,000 steps. This is an expansion of my Snakebite 2.3 dataset with around 700 new images and captions reworked for Anima. Training took approximately 48 hours on a Geforce 3090.
+
+Pros:
+
+    Extremely fast.
+
+    Extremely good prompt adherence.
+
+    Anatomy is pretty stable. If it screws something up, changing your steps by +1/-1 usually fixes it.
+
+    Supports up to nearly 2MP with little-to-no distortions.
+
+At first, I noticed that Photanima's style was inconsistent - it had a tendency to regress toward a cartoony/CGI look as my prompts became more complex. I was able to mostly overcome this by splitting Photanima into constituent content, style-early, and style-late blocks, then boosted the style blocks well past a strength of 1.
+
+"Style-late" maps to blocks 7, 8, and 9 - these do alter composition to a degree, so we can't boost them as hard as "style-early."
+
+Images are pretty consistent now, but there are some notable drawbacks.
+
+Cons in v2:
+
+    It loses a little knowledge of certain artistic terms like silhouette.
+
+    Microdetail quality is somewhere between SDXL and ZIT. Honestly, it's really good for a 2B model. Two-step upscaling with Anima doesn't help much, but I'm sure the results would be amazing if you sent a Photanima image to a different model for refinement. Or if that's too much work: just add a little film grain. It does wonders and requires no extra VRAM.
+
+    Text capabilities are not as good as those of base Anima. Anything beyond 3 or 4 words is likely going to require numerous re-rolls. This is at least partly due to the Turbo LoRA.
+
+    Excessive fluff tags like masterpiece, absurdres, hyperreal tend to fry the image. The model is photographic and highly aesthetic by default, so there's no need to drive it harder in that direction.
+
+🛠️ Recommended Settings (v2)
+
+    6-8 steps.
+
+    er_sde sampler on "ODE" mode.
+
+    Custom sigma curve or simple scheduler: "1, 0.94, 0.9, 0.825, 0.6, 0.5, 0.3, 0.29, 0.2, 0.0"
+
+    CFG 1.
+
+    Preferred resolution: 1040x1520 or 832x1216.
+
+    For maximum realism, begin your prompt with real life photo. If that's not enough, add photo \(medium\) and increase its strength until satisfied. You can usually go up to a crazy strength value like 5 or 6 without breaking the image.
+
+Tips:
+
+    Images often look best at only 6 steps, but anatomy is more stable at 8-10, especially with complex prompts.
+
+    You can reduce the first number on the sigma curve to 0.95-0.99 to improve realism. This reduces saturation and adds a little noise, but makes the model less stable.
+
+Workflows:
+
+    Newest version optimized for realism (recommended): Download
+
+    Simple version with fewer custom nodes: Download
+
+Base model settings:
+
+    Around 40 steps.
+
+    er_sde sampler.
+
+    Simpler scheduler.
+
+    CFG around 4.
+
+    Use a bunch of fluff tags like masterpiece, score_9, absurdres, best quality, highres, photo \(medium\), real life. Note: do not do this with Turbo.