Fix Server API progress signature

app.py

@@ -277,16 +277,13 @@ def _run_search(
     custom_candidates: list[Candidate],
     include_showcase: bool,
     dimension: int,
-    progress: gr.Progress,
 ) -> tuple[torch.Tensor, torch.Tensor, list[Candidate], float, bool]:
     global _SHOWCASE_EMBEDDINGS
 
     started = time.perf_counter()
     was_cold = include_showcase and _SHOWCASE_EMBEDDINGS is None
-    progress(0.04, desc="Allocating the 9B model on GPU")
     MODEL.to("cuda")
     try:
-        progress(0.16, desc=f"Encoding the {query_kind} query")
         query_embedding = _encode([query_payload], query=True)
         document_blocks: list[torch.Tensor] = []
         candidates: list[Candidate] = []
@@ -295,7 +292,6 @@ def _run_search(
             with _CACHE_LOCK:
                 cached = _SHOWCASE_EMBEDDINGS
             if cached is None:
-                progress(0.30, desc="Mapping the curated multimodal universe")
                 encoded = _encode([item.payload for item in SHOWCASE], query=False)
                 with _CACHE_LOCK:
                     if _SHOWCASE_EMBEDDINGS is None:
@@ -305,7 +301,6 @@ def _run_search(
             candidates.extend(SHOWCASE)
 
         if custom_candidates:
-            progress(0.72, desc="Encoding your candidate collection")
             custom_embeddings = _encode([item.payload for item in custom_candidates], query=False)
             document_blocks.append(custom_embeddings)
             candidates.extend(custom_candidates)
@@ -314,7 +309,6 @@ def _run_search(
             raise gr.Error("Include the showcase universe or add at least one candidate.")
 
         documents = torch.cat(document_blocks, dim=0)
-        progress(0.92, desc=f"Ranking in {dimension:,} dimensions")
         return query_embedding, documents, candidates, time.perf_counter() - started, was_cold
     finally:
         MODEL.to("cpu")
@@ -335,15 +329,11 @@ def _pair_duration(*args: Any, **kwargs: Any) -> int:
 def _run_pair(
     query_payload: Any,
     candidate_payload: Any,
-    progress: gr.Progress,
 ) -> tuple[torch.Tensor, torch.Tensor, float]:
     started = time.perf_counter()
-    progress(0.08, desc="Allocating the model on GPU")
     MODEL.to("cuda")
     try:
-        progress(0.35, desc="Encoding the query")
         query = _encode([query_payload], query=True)
-        progress(0.68, desc="Encoding the candidate")
         candidate = _encode([candidate_payload], query=False)
         return query, candidate, time.perf_counter() - started
     finally:
@@ -384,7 +374,6 @@ def search_api(
     candidate_media: list[FileData] | None = None,
     include_showcase: bool = True,
     dimension: int = 1024,
-    progress: gr.Progress = gr.Progress(track_tqdm=True),
 ) -> dict[str, Any]:
     """Search a mixed media collection and return structured visualization data."""
     dimension = int(dimension)
@@ -395,7 +384,7 @@ def search_api(
     custom_candidates = _parse_text_candidates(custom_texts) + _parse_media_candidates(candidate_media)
     with _INFERENCE_LOCK:
         query, documents, candidates, elapsed, was_cold = _run_search(
-            query_payload, query_kind, custom_candidates, bool(include_showcase), dimension, progress,
+            query_payload, query_kind, custom_candidates, bool(include_showcase), dimension,
         )
 
     dimension_map = _dimension_scores(query, documents)
@@ -456,7 +445,6 @@ def compare_api(
     candidate_image: FileData | None = None,
     candidate_video: FileData | None = None,
     dimension: int = 1024,
-    progress: gr.Progress = gr.Progress(track_tqdm=True),
 ) -> dict[str, Any]:
     """Compare two supported inputs and return structured visualization data."""
     dimension = int(dimension)
@@ -468,7 +456,7 @@ def compare_api(
         candidate_text, candidate_image, candidate_video
     )
     with _INFERENCE_LOCK:
-        query, candidate, elapsed = _run_pair(query_payload, candidate_payload, progress)
+        query, candidate, elapsed = _run_pair(query_payload, candidate_payload)
 
     scores = {
         dim: float((_truncate_normalize(query, dim) @ _truncate_normalize(candidate, dim).T).item())