feat(llama.cpp): bump and adapt to upstream changes (#4378)
Some checks are pending
Explorer deployment / build-linux (push) Waiting to run
GPU tests / ubuntu-latest (1.21.x) (push) Waiting to run
generate and publish GRPC docker caches / generate_caches (ubuntu:22.04, linux/amd64,linux/arm64, ubuntu-latest) (push) Waiting to run
generate and publish intel docker caches / generate_caches (intel/oneapi-basekit:2025.0.0-0-devel-ubuntu22.04, linux/amd64, ubuntu-latest) (push) Waiting to run
build container images / hipblas-jobs (-aio-gpu-hipblas, rocm/dev-ubuntu-22.04:6.1, hipblas, true, ubuntu:22.04, extras, latest-gpu-hipblas, latest-aio-gpu-hipblas, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -hipblas-ffmpeg) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas-core) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, false, ubuntu:22.04, extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-intel-f16, quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, true, ubuntu:22.04, extras, latest-gpu-intel-f16, latest-aio-gpu-intel-f16, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -sycl-f16-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-intel-f32, quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, true, ubuntu:22.04, extras, latest-gpu-intel-f32, latest-aio-gpu-intel-f32, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -sycl-f32-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-nvidia-cuda-11, ubuntu:22.04, cublas, 11, 7, true, extras, latest-gpu-nvidia-cuda-11, latest-aio-gpu-nvidia-cuda-11, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -cublas-cuda11-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-nvidia-cuda-12, ubuntu:22.04, cublas, 12, 0, true, extras, latest-gpu-nvidia-cuda-12, latest-aio-gpu-nvidia-cuda-12, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -cublas-cuda12-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f16-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f16-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f32-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f32-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, , , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, ) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, , true, extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, cublas, 11, 7, , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda11) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, cublas, 12, 0, , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda12) (push) Waiting to run
build container images / core-image-build (-aio-cpu, ubuntu:22.04, , true, core, latest-cpu, latest-aio-cpu, --jobs=4 --output-sync=target, linux/amd64,linux/arm64, arc-runner-set, auto, -ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 11, 7, , core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda11-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 11, 7, true, core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda11-ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 12, 0, , core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda12-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 12, 0, true, core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda12-ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, vulkan, true, core, latest-vulkan-ffmpeg-core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, -vulkan-ffmpeg-core) (push) Waiting to run
Security Scan / tests (push) Waiting to run
Tests extras backends / tests-transformers (push) Waiting to run
Tests extras backends / tests-sentencetransformers (push) Waiting to run
Tests extras backends / tests-rerankers (push) Waiting to run
Tests extras backends / tests-diffusers (push) Waiting to run
Tests extras backends / tests-parler-tts (push) Waiting to run
Tests extras backends / tests-openvoice (push) Waiting to run
Tests extras backends / tests-transformers-musicgen (push) Waiting to run
Tests extras backends / tests-vallex (push) Waiting to run
Tests extras backends / tests-coqui (push) Waiting to run
tests / tests-linux (1.21.x) (push) Waiting to run
tests / tests-aio-container (push) Waiting to run
tests / tests-apple (1.21.x) (push) Waiting to run

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-12-14 00:30:52 +01:00 committed by GitHub
parent 0429e00746
commit fc4a714992
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 32 additions and 3 deletions

View File

@ -8,7 +8,7 @@ DETECT_LIBS?=true
# llama.cpp versions
GOLLAMA_REPO?=https://github.com/go-skynet/go-llama.cpp
GOLLAMA_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
CPPLLAMA_VERSION?=274ec65af6e54039eb95cb44904af5c945dca1fa
CPPLLAMA_VERSION?=c27ac678dd393af0da9b8acf10266e760c8a0912
# whisper.cpp version
WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp

View File

@ -2228,6 +2228,35 @@ json parse_options(bool streaming, const backend::PredictOptions* predict, llama
// }
// }
const std::vector<ggml_type> kv_cache_types = {
GGML_TYPE_F32,
GGML_TYPE_F16,
GGML_TYPE_BF16,
GGML_TYPE_Q8_0,
GGML_TYPE_Q4_0,
GGML_TYPE_Q4_1,
GGML_TYPE_IQ4_NL,
GGML_TYPE_Q5_0,
GGML_TYPE_Q5_1,
};
static ggml_type kv_cache_type_from_str(const std::string & s) {
for (const auto & type : kv_cache_types) {
if (ggml_type_name(type) == s) {
return type;
}
}
throw std::runtime_error("Unsupported cache type: " + s);
}
static std::string get_all_kv_cache_types() {
std::ostringstream msg;
for (const auto & type : kv_cache_types) {
msg << ggml_type_name(type) << (&type == &kv_cache_types.back() ? "" : ", ");
}
return msg.str();
}
static void params_parse(const backend::ModelOptions* request,
common_params & params) {
@ -2242,10 +2271,10 @@ static void params_parse(const backend::ModelOptions* request,
// params.model_alias ??
params.model_alias = request->modelfile();
if (!request->cachetypekey().empty()) {
params.cache_type_k = request->cachetypekey();
params.cache_type_k = kv_cache_type_from_str(request->cachetypekey());
}
if (!request->cachetypevalue().empty()) {
params.cache_type_v = request->cachetypevalue();
params.cache_type_v = kv_cache_type_from_str(request->cachetypevalue());
}
params.n_ctx = request->contextsize();
//params.memory_f16 = request->f16memory();