Comment on Mesh LLM: distributed AI computing on irohparentComments−stymaar2moIndeed, I completely screwed my math up. Looks like 10am is too early in the morning for a Sunday.33 tps max token generation speed would be for 10ms of network latency in the above example.
Comments
Indeed, I completely screwed my math up. Looks like 10am is too early in the morning for a Sunday.
33 tps max token generation speed would be for 10ms of network latency in the above example.