coulomb/markitect-main
2
0
Fork 0
Code Issues 62 Pull Requests Actions Projects 12 Releases 1 Wiki Activity

fix(pipeline): retry on all LLM errors (not just rate limits)

Browse Source 
Free-tier APIs intermittently return invalid JSON or empty responses.
Now any exception in _call_llm retries up to 3 times with a 5s back-off,
rather than failing immediately on non-rate-limit errors.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Bernd Worsch
2026-02-19 20:32:23 +01:00
parent 06e904ccf5
commit 1b9a31665c
1 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
markitect/infospace/pipeline.py
View File

@@ -584,9 +584,14 @@ class SourcePipeline:
except Exception as exc: except Exception as exc:
msg = str(exc) msg = str(exc)
print(f" LLM error: {msg}") print(f" LLM error: {msg}")
meta["error"] = msg if attempt < max_retries:
meta["duration_seconds"] = round(time.time() - t0, 1) wait = 5 * (attempt + 1)
return None, meta print(f" Retrying in {wait}s (attempt {attempt + 1}/{max_retries})...")
time.sleep(wait)
else:
meta["error"] = msg
meta["duration_seconds"] = round(time.time() - t0, 1)
return None, meta
if response is None: if response is None:
meta["error"] = "no response" meta["error"] = "no response"