GPT-4.5 vs Grok-2

GPT-4.5 (OpenAI) and Grok-2 (xAI) are both frontier AI models. GPT-4.5 was published in February 2025 and Grok-2 in August 2024.

GPT-4.5 was trained on 3.8 x 10^26 FLOP, about 12.8x the compute of Grok-2 at 3 x 10^25 FLOP. Training compute is the closest available proxy for how much was invested in a model, though it says nothing on its own about how well that compute was spent.

GPT-4.5 wins the single benchmark both were scored on. Benchmark counts are a crude scoreboard — the evaluations differ wildly in what they measure and in how saturated they are — so the per-benchmark table below matters more than the tally.