Could Anthropic have nerfed Opus to boost the perceived advantage of Mythos?
CLAUDE OPUS 4.6 IS NERFED.
BridgeBench just proved it.
Last week Claude Opus 4.6 ranked #2 on the Hallucination benchmark with an accuracy of 83.3%.
Today Claude Opus 4.6 was retested and it fell to #10 on the leaderboard with an accuracy of only 68.3%.
A 98% increase in hallucination.
bridgebench.ai just confirmed that Claude Opus 4.6 has reduced reasoning levels and is nerfed.
Readers added context they thought people might want to know
The cited BridgeBench scores for Claude Opus 4.6 are from tests with different scopes: 6 tasks initially (83.3% accuracy) vs. 30 tasks recently (68.3% accuracy). Analysis of the 6 common tasks shows similar performance (87.6% vs. 85.4%), suggesting no major change.
bridgebench.ai/hallucination/…
bridgebench.ai/hallucination/…
x.com/paul_cal/statu…
PUBLISH THE PROJECT ONLINE
…
START GENERATING MRR FROM IT
USE AI TO BUILD PERSONAL PROJECTS
USE AI TO BUILD PERSONAL PROJECTS
USE AI TO BUILD PERSONAL PROJECTS
USE AI TO BUILD PERSONAL PROJECTS
Unbelievable!
#Deepseek just hit +129% profit.
We need an AI tool that integrates this LLM into trading ASAP!
This return outperforms any hedge fund firm.
