Releases: fairydreaming/llama.cpp
Releases · fairydreaming/llama.cpp
b4255
vulkan: optimize and reenable split_k (#10637) Use vector loads when possible in mul_mat_split_k_reduce. Use split_k when there aren't enough workgroups to fill the shaders.
b4219
sycl : Reroute permuted mul_mats through oneMKL (#10408) This PR fixes the failing MUL_MAT tests for the sycl backend.
b3605
cann: add doc for cann backend (#8867) Co-authored-by: xuedinge233 <[email protected]> Co-authored-by: hipudding <[email protected]>
b3357
Update README.md to fix broken link to docs (#8399) Update the "Performance troubleshooting" doc link to be correct - the file was moved into a dir called 'development'
b3212
disable publishing the full-rocm docker image (#8083)
b3197
vulkan: detect multiple devices by deviceUUID instead of deviceID (#8…
b3190
common: fix warning (#8036) * common: fix warning * Update common/common.cpp Co-authored-by: slaren <[email protected]> --------- Co-authored-by: slaren <[email protected]>
b3145
rpc : fix ggml_backend_rpc_supports_buft() (#7918)
b2963
CUDA: remove incorrect precision check (#7454)