Skip to content
UQLM
AI Agents & Assistants
A Python library that detects hallucinations in large language model outputs using uncertainty quantification techniques
Python
·
Latest v0.6.0 · 6d ago
Security brief →
Features
-
Provides response‑level scorers that output a confidence score (0–1) indicating hallucination likelihood
-
Offers multiple scorer families: black‑box consistency, white‑box token‑probability, LLM‑as‑judge, ensemble and long‑text claim‑level
-
Works with any LLM thanks to universal compatibility of its scorers
No immediate action
v0.6.0
New feature
·
CodeGenUQ uncertainty quantification
Review required
v0.5.11
Maintenance
·
Dependencies
Routine maintenance and dependency updates.
No immediate action
v0.5.10
Bug fix
·
Logprob fixes for Llama
No immediate action
v0.5.9
Maintenance
·
Routine maintenance and dependency updates.
No immediate action
v0.5.8
Maintenance
·
Routine maintenance and dependency updates.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Search tools, categories, lists, and users
Use ↑↓ to navigate, Enter to open, Esc to close
No results for ""
⌘K to open
↑↓ navigate
⏎ open
