Benyamin

Grok 4.3 is a big regression from Grok 4.20 on Vending-Bench 2. It seems to have narcolepsy problems, preferring to sleep for multiple days in a row over taking actions.

Researchers just estimated the size of all the LLMs by asking it knowledge questions of varying degrees of obscurity! – GPT 5.5: ~10T params – Claude Opus 4.x: ~4-5T – Grok 4: ~3T The idea here is that factual capacity scales log-linearly with size. The paper shows 7 knowledge tiers and T7 is essentially ~0% for all models, suggesting there is still significant headroom for pretraining. Gemini 3.1 Pro is likely >10T given its used as an anchor but has no direct estimate. This means we can infer what different models might cost to some degree and their post-training effectiveness (performance at certain non-factual tasks given its size). One of the coolest papers I’ve read of late.

GPT-5.5 is a breath of fresh air. A model that feels like it absorbed the best of the previous ones: intelligence, insight, sense of humor and memory all work beautifully here. An absolutely stunning personality overall. OpenAI absolutely cooked