Gemma 3n: the 4b LLM that’s up with sonnet-3.7 in chatbot arena
the new innovation is Per-Layer Embeddings, which let it consume dramatically less memory
it was created for phones, and is being rolled out to Android phones soon
developers.googleblog.com/en/introduci...
Gemma 3n: the 4b LLM that’s up with sonnet-3.7 in chatbot arena
View original thread
38
3
small note — they call it a 4b but it’s actually an 8b
it uses the memory of a 4b without quantization. i’m not entirely supportive of that naming, but i can see what they were going for
it uses the memory of a 4b without quantization. i’m not entirely supportive of that naming, but i can see what they were going for
7
2 hours later
wait what’s this MatFormer thing?
5