feat(llama.cpp): expose cache_type_k and cache_type_v for quant of kv cache #5353
Triggered via pull request
December 6, 2024 08:14
Status
Cancelled
Total duration
1m 58s
Artifacts
–
Annotations
2 errors
extras-image-build (cublas, 12, 0, linux/amd64, false, -cublas-cuda12-ffmpeg, true, extras, arc-r... / reusable_image-build
Canceling since a higher priority waiting request for 'ci-feat/llama.cpp-quantcache-mudler/LocalAI' exists
|
|