mirror of
https://github.com/mudler/LocalAI.git
synced 2025-05-28 22:44:59 +00:00
feat: move other backends to grpc
This finally makes everything more consistent Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
parent
5dcfdbe51d
commit
1d0ed95a54
54 changed files with 3171 additions and 1712 deletions
|
@ -1,90 +0,0 @@
|
|||
syntax = "proto3";
|
||||
|
||||
option go_package = "github.com/go-skynet/LocalAI/pkg/grpc/proto";
|
||||
option java_multiple_files = true;
|
||||
option java_package = "io.skynet.localai.llmserver";
|
||||
option java_outer_classname = "LLMServer";
|
||||
|
||||
package llm;
|
||||
|
||||
service LLM {
|
||||
rpc Health(HealthMessage) returns (Reply) {}
|
||||
rpc Predict(PredictOptions) returns (Reply) {}
|
||||
rpc LoadModel(ModelOptions) returns (Result) {}
|
||||
rpc PredictStream(PredictOptions) returns (stream Reply) {}
|
||||
rpc Embedding(PredictOptions) returns (EmbeddingResult) {}
|
||||
}
|
||||
|
||||
message HealthMessage {}
|
||||
|
||||
// The request message containing the user's name.
|
||||
message PredictOptions {
|
||||
string Prompt = 1;
|
||||
int32 Seed = 2;
|
||||
int32 Threads = 3;
|
||||
int32 Tokens = 4;
|
||||
int32 TopK = 5;
|
||||
int32 Repeat = 6;
|
||||
int32 Batch = 7;
|
||||
int32 NKeep = 8;
|
||||
float Temperature = 9;
|
||||
float Penalty = 10;
|
||||
bool F16KV = 11;
|
||||
bool DebugMode = 12;
|
||||
repeated string StopPrompts = 13;
|
||||
bool IgnoreEOS = 14;
|
||||
float TailFreeSamplingZ = 15;
|
||||
float TypicalP = 16;
|
||||
float FrequencyPenalty = 17;
|
||||
float PresencePenalty = 18;
|
||||
int32 Mirostat = 19;
|
||||
float MirostatETA = 20;
|
||||
float MirostatTAU = 21;
|
||||
bool PenalizeNL = 22;
|
||||
string LogitBias = 23;
|
||||
bool MLock = 25;
|
||||
bool MMap = 26;
|
||||
bool PromptCacheAll = 27;
|
||||
bool PromptCacheRO = 28;
|
||||
string Grammar = 29;
|
||||
string MainGPU = 30;
|
||||
string TensorSplit = 31;
|
||||
float TopP = 32;
|
||||
string PromptCachePath = 33;
|
||||
bool Debug = 34;
|
||||
repeated int32 EmbeddingTokens = 35;
|
||||
string Embeddings = 36;
|
||||
}
|
||||
|
||||
// The response message containing the result
|
||||
message Reply {
|
||||
string message = 1;
|
||||
}
|
||||
|
||||
message ModelOptions {
|
||||
string Model = 1;
|
||||
int32 ContextSize = 2;
|
||||
int32 Seed = 3;
|
||||
int32 NBatch = 4;
|
||||
bool F16Memory = 5;
|
||||
bool MLock = 6;
|
||||
bool MMap = 7;
|
||||
bool VocabOnly = 8;
|
||||
bool LowVRAM = 9;
|
||||
bool Embeddings = 10;
|
||||
bool NUMA = 11;
|
||||
int32 NGPULayers = 12;
|
||||
string MainGPU = 13;
|
||||
string TensorSplit = 14;
|
||||
int32 Threads = 15;
|
||||
string LibrarySearchPath = 16;
|
||||
}
|
||||
|
||||
message Result {
|
||||
string message = 1;
|
||||
bool success = 2;
|
||||
}
|
||||
|
||||
message EmbeddingResult {
|
||||
repeated float embeddings = 1;
|
||||
}
|
Loading…
Add table
Add a link
Reference in a new issue