2 ms·
Look for positron.ai talks about their tech, they discuss their approach to scaling LLM workloads with their dedicated hardware. It may not be what is done by O
by suspended_state 1y ago
Look for positron.ai talks about their tech, they discuss their approach to scaling LLM workloads with their dedicated hardware. It may not be what is done by OpenAI or other vendors, but you'll get an idea of the underlying problems.