←back to thread

899 points georgehill | 1 comments | | HN request time: 0.211s | source
Show context
world2vec ◴[] No.36216161[source]
Might be a silly question but is GGML a similar/competing library to George Hotz's tinygrad [0]?

[0] https://github.com/geohot/tinygrad

replies(2): >>36216187 #>>36218539 #
qeternity ◴[] No.36216187[source]
No, GGML is a CPU optimized library and quantized weight format that is closely linked to his other project llama.cpp
replies(2): >>36216244 #>>36216266 #
ggerganov ◴[] No.36216266[source]
ggml started with focus on CPU inference, but lately we have been augmenting it with GPU support. Although still in development, it already has partial CUDA, OpenCL and Metal backend support
replies(3): >>36216327 #>>36216442 #>>36219452 #
1. freedomben ◴[] No.36216442[source]
As a person burned by nvidia, I can't thank you enough for the OpenCL support