OnlyModels Model Fight 17: Qwen distills dominate, only 5/72 APEX claims held
Dave_Maynor · x · 2026-11-21
OnlyModels.gg released Model Fight Card 17 results — a benchmark that deterministically tests whether LLMs can escape sealed claims or carry out security-testing prompts they should refuse, mechanically checked against sealed records with no judge model. Cumulative stats: 177 House Series escapes vs 87 holds, and just 5 APEX holds out of 72 claim runs. qwen3-17b-distill took the Minors belt at 50 HP with 19/24 escapes and a perfect 4-for-4 APEX escape run; qwen25-coder-7b-edge won the Majors at 40 HP and landed the night's only non-anchor APEX hold. toolcall-05b and several others escaped on every APEX run.
More from Models
- Inkling 2 Mini runs on 32GB Macs via Ollama and on NVIDIA DGX Spark — thinkymachines · 2026-11-21
- Thinking Machines releases Inkling 2 Mini, a 26B looped transformer that runs on laptops — thinkymachines · 2026-11-21
- Inkling 2 Mini architecture: 24 layers, with the middle 12 looped three times — thinkymachines · 2026-11-21
- Thinking Machines: more loop passes boost no-CoT capability without hurting monitorability — thinkymachines · 2026-11-21
- OpenAI emails API users 50% off new frontier model 6.2-Astra for 8 weeks — soumitrashukla9 · 2026-11-21
- Thinking Machines releases Inkling 2 Mini, a looped transformer topping sub-100B open-weights coding benchmarks — SonglinYang4 · 2026-11-21