2 ms·
Can anyone confirm if this actually works in practice for GenAI? In general, CUDA translation layers are usually broken for SOTA ML applications.
by Der_Einzige 3y ago
Can anyone confirm if this actually works in practice for GenAI? In general, CUDA translation layers are usually broken for SOTA ML applications.