refactor(api): update cli tools for DeepBooru tagger migration

- Update cli/api-interrogate.py to use /sdapi/v1/tagger for DeepBooru
- Handle tagger response format (scores dict or tags string)
- Remove DeepBooru test from interrogate endpoint tests
- Update API model descriptions to reference tagger for anime tagging

cli/api-interrogate.py

@@ -9,6 +9,7 @@ import sys
 import os
 import asyncio
 import filetype
+from types import SimpleNamespace
 from PIL import Image
 from util import log, Map
 import sdapi
@@ -65,22 +66,21 @@ async def interrogate(f):
                 stats['captions'][word] = stats['captions'][word] + 1 if word in stats['captions'] else 1
     else:
         log.error({ 'interrogate clip error': res })
-    # run booru
-    json.model = 'deepdanbooru'
-    res = await sdapi.post('/sdapi/v1/interrogate', json)
+    # run tagger (DeepBooru)
+    tagger_req = SimpleNamespace(image=json.image, model='deepbooru', show_scores=True)
+    res = await sdapi.post('/sdapi/v1/tagger', tagger_req)
     keywords = {}
-    if 'caption' in res:
-        for term in res.caption.split(', '):
-            term = term.replace('(', '').replace(')', '').replace('\\', '').split(':')
-            if len(term) < 2:
-                continue
-            keywords[term[0]] = term[1]
-        keywords = dict(sorted(keywords.items(), key=lambda x:x[1], reverse=True))
-        for word in keywords.items():
-            stats['keywords'][word[0]] = stats['keywords'][word[0]] + 1 if word[0] in stats['keywords'] else 1
-        log.info({ 'interrogate keywords': keywords })
+    if 'scores' in res and res.scores:
+        keywords = dict(sorted(res.scores.items(), key=lambda x: x[1], reverse=True))
+        for word in keywords:
+            stats['keywords'][word] = stats['keywords'][word] + 1 if word in stats['keywords'] else 1
+        log.info({'interrogate keywords': keywords})
+    elif 'tags' in res:
+        for tag in res.tags.split(', '):
+            stats['keywords'][tag] = stats['keywords'][tag] + 1 if tag in stats['keywords'] else 1
+        log.info({'interrogate tags': res.tags})
     else:
-        log.error({ 'interrogate booru error': res })
+        log.error({'interrogate tagger error': res})
     return caption, keywords, style
 
 

cli/test-caption-api.py

@@ -3,7 +3,7 @@
 Caption API Test Suite
 
 Comprehensive tests for all Caption API endpoints and parameters:
-- GET/POST /sdapi/v1/interrogate (OpenCLiP/DeepBooru)
+- GET/POST /sdapi/v1/interrogate (OpenCLiP)
 - POST /sdapi/v1/vqa (VLM Captioning with annotated images)
 - GET /sdapi/v1/vqa/models, /sdapi/v1/vqa/prompts
 - POST /sdapi/v1/tagger
@@ -302,13 +302,7 @@ class CaptionAPITest:
             self.log_fail(f"Expected list, got {type(data)}")
             return
 
-        # Test 2: Contains deepdanbooru
-        if 'deepdanbooru' in data:
-            self.log_pass("Contains 'deepdanbooru'")
-        else:
-            self.log_fail("Missing 'deepdanbooru'")
-
-        # Test 3: Contains OpenCLIP models (format: arch/dataset)
+        # Test 2: Contains OpenCLIP models (format: arch/dataset)
         clip_models = [m for m in data if '/' in m]
         if clip_models:
             self.log_pass(f"Contains {len(clip_models)} OpenCLIP models")
@@ -316,36 +310,6 @@ class CaptionAPITest:
         else:
             self.log_skip("No OpenCLIP models found (may need to download)")
 
-    # =========================================================================
-    # TEST: POST /sdapi/v1/interrogate - DeepBooru
-    # =========================================================================
-    def test_interrogate_post_deepbooru(self):
-        """Test DeepBooru interrogation."""
-        print("\n" + "=" * 70)
-        print("TEST: POST /sdapi/v1/interrogate (DeepBooru)")
-        print("=" * 70)
-
-        t0 = time.time()
-        data = self.post('/sdapi/v1/interrogate', {
-            'image': self.image_b64,
-            'model': 'deepdanbooru'
-        })
-        elapsed = time.time() - t0
-
-        if 'error' in data:
-            self.log_skip(f"DeepBooru: {data.get('reason', 'failed')} (model may not be loaded)")
-            return
-
-        caption = data.get('caption', '')
-        if caption and not self.is_error_answer(caption):
-            caption_preview = caption[:80] + '...' if len(caption) > 80 else caption
-            self.log_pass(f"DeepBooru returns caption ({elapsed:.1f}s)")
-            self.log_info(f"Caption: {caption_preview}")
-        elif self.is_error_answer(caption):
-            self.log_fail(f"DeepBooru returned error: {caption}")
-        else:
-            self.log_fail("DeepBooru returned empty caption")
-
     # =========================================================================
     # TEST: POST /sdapi/v1/interrogate - OpenCLIP Modes
     # =========================================================================
@@ -2323,7 +2287,6 @@ class CaptionAPITest:
 
         # Interrogate tests
         self.test_interrogate_list_models()
-        self.test_interrogate_post_deepbooru()
         self.test_interrogate_post_clip_modes()
         self.test_interrogate_analyze()
         self.test_interrogate_invalid_inputs()

modules/api/endpoints.py

@@ -98,16 +98,17 @@ def get_interrogate():
     Returns model identifiers for use with POST /sdapi/v1/interrogate.
 
     **Model Types:**
-    - `deepdanbooru`: Anime-style image tagger returning comma-separated tags
     - OpenCLIP models: Format `architecture/pretrained_dataset` (e.g., `ViT-L-14/openai`)
 
+    For anime-style tagging (WaifuDiffusion, DeepBooru), use `/sdapi/v1/tagger` instead.
+
     **Example Response:**
     ```json
-    ["deepdanbooru", "ViT-L-14/openai", "ViT-H-14/laion2b_s32b_b79k"]
+    ["ViT-L-14/openai", "ViT-H-14/laion2b_s32b_b79k"]
     ```
     """
     from modules.interrogate.openclip import refresh_clip_models
-    return ['deepdanbooru'] + refresh_clip_models()
+    return refresh_clip_models()
 
 def get_schedulers():
     from modules.sd_samplers import list_samplers

modules/api/models.py

@@ -368,11 +368,11 @@ class ResStatus(BaseModel):
 class ReqInterrogate(BaseModel):
     """Request model for OpenCLIP/BLIP image interrogation.
 
-    Analyze image using CLIP model via OpenCLIP to generate prompts,
-    or use DeepDanbooru for anime-style tagging.
+    Analyze image using CLIP model via OpenCLIP to generate prompts.
+    For anime-style tagging, use /sdapi/v1/tagger with WaifuDiffusion or DeepBooru.
     """
     image: str = Field(default="", title="Image", description="Image to interrogate. Must be a Base64 encoded string containing the image data (PNG/JPEG).")
-    model: str = Field(default="ViT-L-14/openai", title="Model", description="OpenCLIP model to use. Use 'deepdanbooru' or 'deepbooru' for anime tagging. Get available models from GET /sdapi/v1/interrogate.")
+    model: str = Field(default="ViT-L-14/openai", title="Model", description="OpenCLIP model to use. Get available models from GET /sdapi/v1/interrogate.")
     clip_model: str = Field(default="ViT-L-14/openai", title="CLIP Model", description="CLIP model used for image-text similarity matching. Larger models (ViT-L, ViT-H) are more accurate but slower and use more VRAM.")
     blip_model: str = Field(default="blip-large", title="Caption Model", description="BLIP model used to generate the initial image caption. The caption model describes the image content which CLIP then enriches with style and flavor terms.")
     mode: str = Field(default="best", title="Mode", description="Interrogation mode. Fast: Quick caption with minimal flavor terms. Classic: Standard interrogation with balanced quality and speed. Best: Most thorough analysis, slowest but highest quality. Negative: Generate terms to use as negative prompt.")