3 points | by faangguyindia 3 hours ago
2 comments
That's awesome I came from the jev in python thread. is the model shareable? do you have any more writing on this? would love to read more about it! cool demo anyways.
Here you can see a 26B model hacked using a trained KV cache bank, responding like Gemma. The latency I am seeing is <127 ms.
That's awesome I came from the jev in python thread. is the model shareable? do you have any more writing on this? would love to read more about it! cool demo anyways.
Here you can see a 26B model hacked using a trained KV cache bank, responding like Gemma. The latency I am seeing is <127 ms.