llama_cpp 0.15.1 → 0.15.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +14 -0
- data/ext/llama_cpp/llama_cpp.cpp +49 -0
- data/lib/llama_cpp/version.rb +2 -2
- data/sig/llama_cpp.rbs +4 -0
- data/vendor/tmp/llama.cpp/Makefile +9 -20
- data/vendor/tmp/llama.cpp/ggml-backend.c +2 -3
- data/vendor/tmp/llama.cpp/ggml-common.h +0 -54
- data/vendor/tmp/llama.cpp/ggml-cuda.cu +87 -37
- data/vendor/tmp/llama.cpp/ggml-cuda.h +1 -0
- data/vendor/tmp/llama.cpp/ggml-impl.h +47 -0
- data/vendor/tmp/llama.cpp/ggml-kompute.cpp +13 -3
- data/vendor/tmp/llama.cpp/ggml-metal.m +177 -190
- data/vendor/tmp/llama.cpp/ggml-metal.metal +97 -505
- data/vendor/tmp/llama.cpp/ggml-opencl.cpp +4 -1
- data/vendor/tmp/llama.cpp/ggml-quants.c +3660 -2057
- data/vendor/tmp/llama.cpp/ggml-rpc.cpp +1155 -0
- data/vendor/tmp/llama.cpp/ggml-rpc.h +24 -0
- data/vendor/tmp/llama.cpp/ggml-sycl.cpp +60 -639
- data/vendor/tmp/llama.cpp/ggml-vulkan-shaders.hpp +9351 -5627
- data/vendor/tmp/llama.cpp/ggml-vulkan.cpp +203 -224
- data/vendor/tmp/llama.cpp/ggml.c +1168 -1470
- data/vendor/tmp/llama.cpp/ggml.h +67 -44
- data/vendor/tmp/llama.cpp/llama.cpp +1371 -944
- data/vendor/tmp/llama.cpp/llama.h +13 -3
- data/vendor/tmp/llama.cpp/unicode-data.cpp +6969 -2169
- data/vendor/tmp/llama.cpp/unicode-data.h +15 -12
- data/vendor/tmp/llama.cpp/unicode.cpp +89 -111
- data/vendor/tmp/llama.cpp/unicode.h +44 -12
- metadata +5 -3
|
@@ -0,0 +1,24 @@
|
|
|
1
|
+
#pragma once
|
|
2
|
+
|
|
3
|
+
#include "ggml.h"
|
|
4
|
+
#include "ggml-backend.h"
|
|
5
|
+
|
|
6
|
+
#ifdef __cplusplus
|
|
7
|
+
extern "C" {
|
|
8
|
+
#endif
|
|
9
|
+
|
|
10
|
+
#define GGML_RPC_MAX_SERVERS 16
|
|
11
|
+
|
|
12
|
+
// backend API
|
|
13
|
+
GGML_API GGML_CALL ggml_backend_t ggml_backend_rpc_init(const char * endpoint);
|
|
14
|
+
GGML_API GGML_CALL bool ggml_backend_is_rpc(ggml_backend_t backend);
|
|
15
|
+
|
|
16
|
+
GGML_API GGML_CALL ggml_backend_buffer_type_t ggml_backend_rpc_buffer_type(const char * endpoint);
|
|
17
|
+
|
|
18
|
+
GGML_API GGML_CALL void ggml_backend_rpc_get_device_memory(const char * endpoint, size_t * free, size_t * total);
|
|
19
|
+
|
|
20
|
+
GGML_API GGML_CALL void start_rpc_server(ggml_backend_t backend, const char * endpoint, size_t free_mem, size_t total_mem);
|
|
21
|
+
|
|
22
|
+
#ifdef __cplusplus
|
|
23
|
+
}
|
|
24
|
+
#endif
|