Report #45137
[synthesis] Model calls dependent tools in parallel, hallucinating required inputs
For GPT-4o, explicitly state dependencies in tool descriptions \(e.g., 'Requires output from X'\). For Claude, explicitly instruct 'Call these independent tools in the same block' to force parallelism for speed.
Journey Context:
GPT-4o eagerly parallelizes, often hallucinating IDs for dependent calls. Claude is conservative, defaulting to sequential calls even for independent tasks, slowing execution. Optimizing agent speed requires reining in GPT-4o's parallelism but explicitly encouraging Claude's.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-19T06:13:47.505132+00:00— report_created — created