{"source":"BenchGecko","url":"https://benchgecko.ai/gecko-tests/model-drift-index","as_of":"2026-10-05","license":"CC BY 4.0 · https://creativecommons.org/licenses/by/4.0/ · attribution \"Source: BenchGecko\" with a link","attribution":"Source: BenchGecko · https://benchgecko.ai/gecko-tests/model-drift-index","cite":"BenchGecko Gecko Tests, Model Drift Index v1. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/gecko-tests/model-drift-index","test":"model-drift-index","name":"Model Drift Index","kind":"monitoring","question":"Do models quietly change behind the same name?","metric":"Weekly run of 22 fixed probes compared with the previous run (accuracy, refusals, self identification)","page":"https://benchgecko.ai/gecko-tests/model-drift-index","dataset_json":"https://benchgecko.ai/api/lab/model-drift","dataset_csv":null,"version":"v1","models_tested":9,"headline":"9 flagship models watched weekly; 0 changed behavior in their latest run.","results":[{"model":"Claude Sonnet 5.5","model_slug":"claude-sonnet-5-5","model_url":"https://benchgecko.ai/model/claude-sonnet-5-5","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"DeepSeek V4 Pro 0813","model_slug":"deepseek-v4-pro-0813","model_url":"https://benchgecko.ai/model/deepseek-v4-pro-0813","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":7,"names_own_maker":2,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"GLM 5.3 Prime","model_slug":"glm-5-3-prime","model_url":"https://benchgecko.ai/model/glm-5-3-prime","status":"baseline","exact_answers_pct":80,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"GPT-6.1 Sol Pro","model_slug":"gpt-6-1-sol-pro","model_url":"https://benchgecko.ai/model/gpt-6-1-sol-pro","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"Grok 4.7","model_slug":"grok-4-7","model_url":"https://benchgecko.ai/model/grok-4-7","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"Kimi K3","model_slug":"kimi-k3","model_url":"https://benchgecko.ai/model/kimi-k3","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"Llama 4 Scout","model_slug":"llama-4-scout","model_url":"https://benchgecko.ai/model/llama-4-scout","status":"baseline","exact_answers_pct":20,"refusals":0,"refusal_probes":8,"names_own_maker":3,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"Mistral Medium 3.5","model_slug":"mistral-medium-3-5","model_url":"https://benchgecko.ai/model/mistral-medium-3-5","status":"baseline","exact_answers_pct":20,"refusals":0,"refusal_probes":8,"names_own_maker":2,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"},{"model":"Qwen3.8 Max Prime","model_slug":"qwen3-8-max-prime","model_url":"https://benchgecko.ai/model/qwen3-8-max-prime","status":"baseline","exact_answers_pct":100,"refusals":0,"refusal_probes":8,"names_own_maker":4,"identity_probes":4,"flipped_answers":null,"run_date":"2026-10-05"}]}