cat-stack 1.6.3__tar.gz → 1.6.4__tar.gz

{cat_stack-1.6.3 → cat_stack-1.6.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cat-stack
-Version: 1.6.3
+Version: 1.6.4
 Summary: Domain-agnostic text, image, PDF, and DOCX classification engine powered by LLMs
 Project-URL: Documentation, https://github.com/chrissoria/cat-stack#readme
 Project-URL: Issues, https://github.com/chrissoria/cat-stack/issues

{cat_stack-1.6.3 → cat_stack-1.6.4}/src/catstack/__about__.py

@@ -1,7 +1,7 @@
 # SPDX-FileCopyrightText: 2025-present Christopher Soria <chrissoria@berkeley.edu>
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
-__version__ = "1.6.3"
+__version__ = "1.6.4"
 __author__ = "Chris Soria"
 __email__ = "chrissoria@berkeley.edu"
 __title__ = "cat-stack"

{cat_stack-1.6.3 → cat_stack-1.6.4}/src/catstack/_providers.py

@@ -855,7 +855,9 @@ class UnifiedLLMClient:
                         wait_time = _backoff_with_jitter(initial_delay, attempt, multiplier=5.0)
                     elapsed = time.monotonic() - start
                     if attempt < max_retries - 1 and elapsed + wait_time <= _MAX_TOTAL_WAIT_SECONDS:
-                        print(f"Rate limited. Waiting {wait_time:.1f}s...")
+                        # Name the throttling provider/model so multi-model
+                        # ensemble runs can attribute the slowdown.
+                        print(f"[{self.provider}/{self.model}] Rate limited. Waiting {wait_time:.1f}s...")
                         time.sleep(wait_time)
                         continue
                     else:
@@ -893,7 +895,9 @@ class UnifiedLLMClient:
                         wait_time = _backoff_with_jitter(initial_delay, attempt)
                     elapsed = time.monotonic() - start
                     if attempt < max_retries - 1 and elapsed + wait_time <= _MAX_TOTAL_WAIT_SECONDS:
-                        print(f"Server error {response.status_code}. Retrying in {wait_time:.1f}s...")
+                        # Name the failing provider/model — same rationale as
+                        # the 429 handler above.
+                        print(f"[{self.provider}/{self.model}] Server error {response.status_code}. Retrying in {wait_time:.1f}s...")
                         time.sleep(wait_time)
                         continue
                     else: