CUDA: faster dequantize kernels for Q4_0 and Q4_1 (#4938) #16
Annotations
2 errors
|
|
|
Build and push Docker image (versioned)
buildx failed with: ERROR: invalid tag "ghcr.io/YannFollet/llama.cpp:light-rocm-4a3156de2fac9a8ee4279de7804d4e352dcfe121": repository name must be lowercase
|
The logs for this run have expired and are no longer available.
Loading