Comment by wavemode 1 month ago He tests several Claude versions as well 1 comment wavemode Reply causal 1 month ago Ah you're right, scrolled past that - the most salient contrast in the chart is still just GPT-5 vs GPT-4, and it feels easy to contrive such results by pinning one model's response as "ideal" and making that a benchmark for everything else.
causal 1 month ago Ah you're right, scrolled past that - the most salient contrast in the chart is still just GPT-5 vs GPT-4, and it feels easy to contrive such results by pinning one model's response as "ideal" and making that a benchmark for everything else.
Ah you're right, scrolled past that - the most salient contrast in the chart is still just GPT-5 vs GPT-4, and it feels easy to contrive such results by pinning one model's response as "ideal" and making that a benchmark for everything else.