1 points | by behnamoh 4 hours ago
1 comments
However, I think it lacks the most interesting information - which is a latency in tokens/sec and how it decays with increasing of parallelism. I also not exactly got what it has to do with AI Agents in particular.
However, I think it lacks the most interesting information - which is a latency in tokens/sec and how it decays with increasing of parallelism. I also not exactly got what it has to do with AI Agents in particular.