{"version": 1, "date": "2026-10-29", "taskSet": "2026-10-29.1", "tasks": 25, "runsPerTask": 1, "temperature": 0.0, "previousDate": null, "finding": "GPT-4.1 answered 96% of 25 checkable tasks correctly; GPT-4o mini 92% at 13 times less.", "categories": ["dates-units", "extraction", "instructions", "knowledge", "reasoning"], "models": [{"model": "gpt-4.1", "label": "GPT-4.1", "where": "OpenAI API", "accuracy": 96.0, "accuracyChange": null, "correct": 24, "errors": 0, "latencyP50Seconds": 0.83, "costPerRunUsd": 0.002842, "costPer100TasksUsd": 0.011368, "tokens": 1007, "consistency": null, "byCategory": {"dates-units": 75.0, "extraction": 100.0, "instructions": 100.0, "knowledge": 100.0, "reasoning": 100.0}, "priceSource": "openai.com/api/pricing (list prices; recheck before publishing)"}, {"model": "gpt-4o-mini", "label": "GPT-4o mini", "where": "OpenAI API", "accuracy": 92.0, "accuracyChange": null, "correct": 23, "errors": 0, "latencyP50Seconds": 0.779, "costPerRunUsd": 0.000218, "costPer100TasksUsd": 0.000872, "tokens": 1015, "consistency": null, "byCategory": {"dates-units": 75.0, "extraction": 100.0, "instructions": 80.0, "knowledge": 100.0, "reasoning": 100.0}, "priceSource": "openai.com/api/pricing (list prices; recheck before publishing)"}, {"model": "gpt-4.1-mini", "label": "GPT-4.1 mini", "where": "OpenAI API", "accuracy": 88.0, "accuracyChange": null, "correct": 22, "errors": 0, "latencyP50Seconds": 0.764, "costPerRunUsd": 0.000575, "costPer100TasksUsd": 0.002299, "tokens": 1011, "consistency": null, "byCategory": {"dates-units": 75.0, "extraction": 100.0, "instructions": 60.0, "knowledge": 100.0, "reasoning": 100.0}, "priceSource": "openai.com/api/pricing (list prices; recheck before publishing)"}], "notReached": [], "missedByAll": [{"task": "date-days", "category": "dates-units", "prompt": "How many days are there from 1 March 2024 to 1 March 2025? Answer with the number only.", "expected": 365, "answers": {"gpt-4.1": "366", "gpt-4.1-mini": "366", "gpt-4o-mini": "366"}}], "sources": {"results": "C:\\Users\\AntiKali\\Documents\\Harness\\ModelBench\\2026-10-29\\results.csv", "methodology": "C:\\Users\\AntiKali\\Documents\\Harness\\ModelBench\\2026-10-29\\methodology.md"}}