MQ
MODELQUEST
choose the right AI for the quest
Claude Sonnet 5 · API · High effort (default)
Evidenceacceptedindependent_lab

AA-Omniscience

Artificial Analysis — AA-Omniscience (Claude Sonnet 5 max)

ev_2026_07_sonnet5_aa_omniscience

Result
Configuration

How it was measured

Exact model
claude-sonnet-5
Reasoning
max
Agent harness
none
Tool access
AA default knowledge probe (not web-search agent)
Prompting
Artificial Analysis Omniscience protocol

PRIMARY independent WIS signal — calibration / uncertainty management via AA-Omniscience (accuracy vs hallucination tradeoff).

Caveats

Known limitations

  • Omniscience Index is not percent accuracy; display accuracy + hallucination rate.
  • Max effort config.
  • Positive index is relatively strong in a field where many models score ≤0.
Weighting

Evidence weight factors

The engine weights each evidence row by these factors (0–1) when fusing scores — higher is more trusted.

relevance0.95
quality1.00
recency1.00
health0.85
comparability0.90
reliability0.85
Benchmark

AA-Omniscience

Knowledge reliability benchmark that jointly scores accuracy and hallucination / overconfidence. Rewards calibrated abstention rather than pure guess accuracy.

Benchmark source →