Knowledge Base Regulation Tracker Standards & Frameworks OSS Directory AI Incident Database Research Paper Library Case Studies Society Articles News Jobs About Charter Sign inBecome a Member
Home/AI Incident Database/"Reflection 70B" claims of beating GPT-4o collapse as researchers find the demo routes to Claude

"Reflection 70B" claims of beating GPT-4o collapse as researchers find the demo routes to Claude

A model announced as the world's best open LLM turned out to underperform, and its hosted demo was found to be calling Anthropic's Claude, damaging trust in self-reported benchmarks.

Severity: Lowreflection 70bbenchmarksfraud

Contribute

Spot an error, or know a record we should add? Members can propose additions and submit corrections.