Refusal-removed Qwen2.5-72B runs on Mac, zero refusals with high risks
petrusenko_max · x · 2026-08-20
A "refusal-removed" version of Qwen2.5-72B is now running locally on Apple Silicon. Released as MLX builds in 2, 4, 6, and 8-bit variants, tests show zero refusals in 4/6/8-bit modes while preserving vision, reasoning, and tool-calling capabilities across a 262K-token context. The creators warn that the model can generate malware, fraud schemes, and weapon instructions on demand, highlighting tangible immediate dangers despite its high capabilities.
More from Models
- Model at capacity for a straight week drives frustrated dev back to Opus — joonasvirtanen · 2026-10-16
- Decision models match LLMs on routing and safety — and are 7-8x faster — randal_olson · 2026-10-16
- Xiaomi MiMo V2.6 Pro takes #1 open-model spot on Agent Arena at $0.11/task — arena · 2026-10-16
- ChatGPT's 'best AI video tool' answer points to Runway integration, sparking ad doubts — tlakomy · 2026-10-16
- Microsoft's MAI finishes in-house 3T-parameter pretrained base models, post-training results coming — soumitrashukla9 · 2026-10-16
- Hands-on with Gemini 4 Argon: faster than Opus 5.5 but hallucinates more and costs more — prasenx · 2026-10-16