4 ms·
Go look at their past blog posts. OpenAI only ever benchmarks against their own models.
by poormathskills 1y ago
Go look at their past blog posts. OpenAI only ever benchmarks against their own models.
- oofbaroomf 1y agoOh, ok. But it's still quite telling of their attitude as an organization.
- rvnx 1y agoIt's the same organization that kept repeating that sharing weights of GPT would be "too dangerous for the world". Eventually DeepSeek thankfully did something like that, though they are supposed to be the evil guys.