Re:RasaGPT: First headless LLM chatbot built on top of Rasa,...
by KaoruAoiShiho on Monday May 08, 2023 @05:46PM
(#35867431)
The claim of 70% of 4090 is very strange, my 4090 runs a 30b at roughly 25 tokens/second compared to the 1token/second claimed by the p40 user here: https://news.ycombinator.com/item?id=35861360
Re:RasaGPT: First headless LLM chatbot built on top of Rasa,...
by MacsHeadroom on Monday May 08, 2023 @10:33PM
(#35869571)
>compared to the 1token/second claimed by the p40 user
That user is doing something wrong. They may not be cooling it and are getting thermal throttled. That would be my guess.
The P40 is capable of upwards of 10 tokens/second with 30b.
I understand that 4090 is aimed at gaming and has a lot of extra bells and whistles like the RTX cores. But it is also consumer electronics and much cheaper than the enterprise GPU lines for the same power.
According to this 4090 already has double the raw flop performance of the V100 and is competitive with the most powerful GPUs in the market from last year. https://www.aime.info/blog/en/deep-learning-gpu-benchmarks-2...
And according to this the V100 is ~60% faster than the P40 https://ai-benchmark.com/ranking_deeplearning_detailed.html