ExternalVendorCode/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2024-12-18 20:27:57 +00:00
Code Issues Actions 12 Packages Projects Releases Wiki Activity

build(Makefile): add back single target to build native llama-cpp (#2448)

Browse Source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-05-31 18:35:33 +02:00 committed by GitHub
parent 10c64dbb55
commit ff8a6962cd
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 17 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
Makefile
View File

@ -672,6 +672,14 @@ else
LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/${VARIANT} grpc-server LLAMA_VERSION=$(CPPLLAMA_VERSION) $(MAKE) -C backend/cpp/${VARIANT} grpc-server
endif endif
# This target is for manually building a variant with-auto detected flags
backend-assets/grpc/llama-cpp: backend-assets/grpc
cp -rf backend/cpp/llama backend/cpp/llama-cpp
$(MAKE) -C backend/cpp/llama-cpp purge
$(info ${GREEN}I llama-cpp build info:avx2${RESET})
$(MAKE) VARIANT="llama-cpp" build-llama-cpp-grpc-server
cp -rfv backend/cpp/llama-cpp/grpc-server backend-assets/grpc/llama-cpp
backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc backend-assets/grpc/llama-cpp-avx2: backend-assets/grpc
cp -rf backend/cpp/llama backend/cpp/llama-avx2 cp -rf backend/cpp/llama backend/cpp/llama-avx2
$(MAKE) -C backend/cpp/llama-avx2 purge $(MAKE) -C backend/cpp/llama-avx2 purge

13
docs/content/docs/advanced/advanced-usage.md
View File

@ -351,7 +351,7 @@ For example, to start vllm manually after compiling LocalAI (also assuming runni
./local-ai --external-grpc-backends "vllm:$PWD/backend/python/vllm/run.sh" ./local-ai --external-grpc-backends "vllm:$PWD/backend/python/vllm/run.sh"
``` ```
Note that first is is necessary to create the conda environment with: Note that first is is necessary to create the environment with:
```bash ```bash
make -C backend/python/vllm make -C backend/python/vllm
@ -369,7 +369,7 @@ there are additional environment variables available that modify the behavior of
| `BUILD_TYPE` | | Build type. Available: `cublas`, `openblas`, `clblas` | | `BUILD_TYPE` | | Build type. Available: `cublas`, `openblas`, `clblas` |
| `GO_TAGS` | | Go tags. Available: `stablediffusion` | | `GO_TAGS` | | Go tags. Available: `stablediffusion` |
| `HUGGINGFACEHUB_API_TOKEN` | | Special token for interacting with HuggingFace Inference API, required only when using the `langchain-huggingface` backend | | `HUGGINGFACEHUB_API_TOKEN` | | Special token for interacting with HuggingFace Inference API, required only when using the `langchain-huggingface` backend |
| `EXTRA_BACKENDS` | | A space separated list of backends to prepare. For example `EXTRA_BACKENDS="backend/python/diffusers backend/python/transformers"` prepares the conda environment on start | | `EXTRA_BACKENDS` | | A space separated list of backends to prepare. For example `EXTRA_BACKENDS="backend/python/diffusers backend/python/transformers"` prepares the python environment on start |
| `DISABLE_AUTODETECT` | `false` | Disable autodetect of CPU flagset on start | | `DISABLE_AUTODETECT` | `false` | Disable autodetect of CPU flagset on start |
| `LLAMACPP_GRPC_SERVERS` | | A list of llama.cpp workers to distribute the workload. For example `LLAMACPP_GRPC_SERVERS="address1:port,address2:port"` | | `LLAMACPP_GRPC_SERVERS` | | A list of llama.cpp workers to distribute the workload. For example `LLAMACPP_GRPC_SERVERS="address1:port,address2:port"` |
@ -475,7 +475,7 @@ If you wish to build a custom container image with extra backends, you can use t
```Dockerfile ```Dockerfile
FROM quay.io/go-skynet/local-ai:master-ffmpeg-core FROM quay.io/go-skynet/local-ai:master-ffmpeg-core
RUN PATH=$PATH:/opt/conda/bin make -C backend/python/diffusers RUN make -C backend/python/diffusers
``` ```
Remember also to set the `EXTERNAL_GRPC_BACKENDS` environment variable (or `--external-grpc-backends` as CLI flag) to point to the backends you are using (`EXTERNAL_GRPC_BACKENDS="backend_name:/path/to/backend"`), for example with diffusers: Remember also to set the `EXTERNAL_GRPC_BACKENDS` environment variable (or `--external-grpc-backends` as CLI flag) to point to the backends you are using (`EXTERNAL_GRPC_BACKENDS="backend_name:/path/to/backend"`), for example with diffusers:
@ -483,7 +483,7 @@ Remember also to set the `EXTERNAL_GRPC_BACKENDS` environment variable (or `--ex
```Dockerfile ```Dockerfile
FROM quay.io/go-skynet/local-ai:master-ffmpeg-core FROM quay.io/go-skynet/local-ai:master-ffmpeg-core
RUN PATH=$PATH:/opt/conda/bin make -C backend/python/diffusers RUN make -C backend/python/diffusers
ENV EXTERNAL_GRPC_BACKENDS="diffusers:/build/backend/python/diffusers/run.sh" ENV EXTERNAL_GRPC_BACKENDS="diffusers:/build/backend/python/diffusers/run.sh"
``` ```
@ -525,3 +525,8 @@ A list of the environment variable that tweaks parallelism is the following:
Note that, for llama.cpp you need to set accordingly `LLAMACPP_PARALLEL` to the number of parallel processes your GPU/CPU can handle. For python-based backends (like vLLM) you can set `PYTHON_GRPC_MAX_WORKERS` to the number of parallel requests. Note that, for llama.cpp you need to set accordingly `LLAMACPP_PARALLEL` to the number of parallel processes your GPU/CPU can handle. For python-based backends (like vLLM) you can set `PYTHON_GRPC_MAX_WORKERS` to the number of parallel requests.
### Disable CPU flagset auto detection in llama.cpp
LocalAI will automatically discover the CPU flagset available in your host and will use the most optimized version of the backends.
If you want to disable this behavior, you can set `DISABLE_AUTODETECT` to `true` in the environment variables.