انویدیا یه مقاله جدید داده بیرون که می‌گه واسه پیدا کردن بهترین مدل زبانی پایه برای ایجنت‌های برنامه‌نویسی، نباید خودت رو درگیر ارزیابی تک‌تک قدم‌های کار کنی

New Nvidia paper shows scoring base models on every step of an agent run works worse than scoring them only on the edit that first makes the tests pass. Post-training is expensive, so picking the right base model matters. Yet agent benchmarks score base models near zero, and simple coding tests barely predict later agent skill. They replay a strong agent's successful run and rerun the tests after each edit to find the 1st edit that makes them pass. At this decisive step they test whether a base model favors that edit, picks it over failing edits, and can write a passing one. How likely each base model was to write that exact edit ranked the same 10 models in line with their post-trained SWE-bench Verified scores, at 0.964. When choosing a base model to post-train for agent work, run this check first. It needs only a benchmark's successful agent runs and its tests.