🔥 add LaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types (#1254)

* wip

* wip

* Make it functional

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* wip

* Small fixups

* do not inject space on role encoding, encode img at beginning of messages

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Add examples/config defaults

* Add include dir of current source dir

* cleanup

* fixes

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* fixups

* Revert "fixups"

This reverts commit f1a4731cca.

* fixes

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

backend/cpp/llama/CMakeLists.txt

@@ -1,9 +1,22 @@
+
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+set(TARGET myclip)
+add_library(${TARGET} clip.cpp clip.h)
+install(TARGETS ${TARGET} LIBRARY)
+target_link_libraries(${TARGET} PRIVATE common ggml ${CMAKE_THREAD_LIBS_INIT})
+target_compile_features(${TARGET} PRIVATE cxx_std_11)
+if (NOT MSVC)
+    target_compile_options(${TARGET} PRIVATE -Wno-cast-qual) # stb_image.h
+endif()
+
+set(TARGET grpc-server)
+# END CLIP hack
 set(CMAKE_CXX_STANDARD 17)
 cmake_minimum_required(VERSION 3.15)
 set(TARGET grpc-server)
 set(_PROTOBUF_LIBPROTOBUF libprotobuf)
 set(_REFLECTION grpc++_reflection)
-
 if (${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
     link_directories("/opt/homebrew/lib")
     include_directories("/opt/homebrew/include")
@@ -47,10 +60,10 @@ add_library(hw_grpc_proto
   ${hw_grpc_srcs}
   ${hw_grpc_hdrs}
   ${hw_proto_srcs}
-  ${hw_proto_hdrs})
+  ${hw_proto_hdrs} )
 
-add_executable(${TARGET} grpc-server.cpp)
-target_link_libraries(${TARGET} PRIVATE common llama ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
+add_executable(${TARGET} grpc-server.cpp json.hpp )
+target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
   absl::flags_parse
   gRPC::${_REFLECTION}
   gRPC::${_GRPC_GRPCPP}

backend/cpp/llama/Makefile

@@ -1,5 +1,5 @@
 
-LLAMA_VERSION?=
+LLAMA_VERSION?=d9b33fe95bd257b36c84ee5769cc048230067d6f
 
 CMAKE_ARGS?=
 BUILD_TYPE?=
@@ -27,11 +27,17 @@ llama.cpp/examples/grpc-server:
 	mkdir -p llama.cpp/examples/grpc-server
 	cp -r $(abspath ./)/CMakeLists.txt llama.cpp/examples/grpc-server/
 	cp -r $(abspath ./)/grpc-server.cpp llama.cpp/examples/grpc-server/
+	cp -rfv $(abspath ./)/json.hpp llama.cpp/examples/grpc-server/
 	echo "add_subdirectory(grpc-server)" >> llama.cpp/examples/CMakeLists.txt
+## XXX: In some versions of CMake clip wasn't being built before llama.
+## This is an hack for now, but it should be fixed in the future.
+	cp -rfv llama.cpp/examples/llava/clip.h llama.cpp/examples/grpc-server/clip.h
+	cp -rfv llama.cpp/examples/llava/clip.cpp llama.cpp/examples/grpc-server/clip.cpp
 
 rebuild:
 	cp -rfv $(abspath ./)/CMakeLists.txt llama.cpp/examples/grpc-server/
 	cp -rfv $(abspath ./)/grpc-server.cpp llama.cpp/examples/grpc-server/
+	cp -rfv $(abspath ./)/json.hpp llama.cpp/examples/grpc-server/
 	rm -rf grpc-server
 	$(MAKE) grpc-server