A review of five uncensored Qwen3.5-9B models indicates that even with modifications, one model still refuses 71% of prompts it claims to handle. The benchmark was conducted on an M5 Mac using 115 prompts per model, evaluating their local performance. AI
IMPACT Highlights potential limitations in uncensored models for local deployment.
RANK_REASON Analysis of model performance on specific hardware. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →