3 ms·
It takes a lot of GPU to run the larger FOSS LLMs. The OpenAI API is still the cheapest way to get high-quality LLM generations.
by dtquad 2y ago
It takes a lot of GPU to run the larger FOSS LLMs. The OpenAI API is still the cheapest way to get high-quality LLM generations.
- ein0p 2y agoTell that to someone else. I run LLaMA70B in my garage. Works fine, cost is negligible.
- ComplexSystems 2y agoWhat, are nation states trying to interfere in US elections on a real tight budget or something?