dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-20 02:24:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 21

feat: add machine tag and inference timings (#4577)

Browse source 
* Add machine tag option, add extraUsage option, grpc-server -> proto -> endpoint extraUsage data is broken for now

Signed-off-by: mintyleaf <mintyleafdev@gmail.com>

* remove redurant timing fields, fix not working timings output

Signed-off-by: mintyleaf <mintyleafdev@gmail.com>

* use middleware for Machine-Tag only if tag is specified

Signed-off-by: mintyleaf <mintyleafdev@gmail.com>

---------

Signed-off-by: mintyleaf <mintyleafdev@gmail.com>
This commit is contained in:
mintyleaf 2025-01-17 20:05:58 +04:00 • committed by GitHub
parent 8027fdf1c7
commit 96f8ec0402
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
15 changed files with 137 additions and 48 deletions
Download patch file Download diff file Expand all files Collapse all files

4
backend/backend.proto
View file

@ -159,6 +159,8 @@ message Reply {
bytes message = 1;
int32 tokens = 2;
int32 prompt_tokens = 3;
double timing_prompt_processing = 4;
double timing_token_generation = 5;
}
message ModelOptions {
@ -348,4 +350,4 @@ message StatusResponse {
message Message {
string role = 1;
string content = 2;
}
}

14
backend/cpp/llama/grpc-server.cpp
View file

@ -2408,6 +2408,13 @@ public:
int32_t tokens_evaluated = result.result_json.value("tokens_evaluated", 0);
reply.set_prompt_tokens(tokens_evaluated);
if (result.result_json.contains("timings")) {
double timing_prompt_processing = result.result_json.at("timings").value("prompt_ms", 0.0);
reply.set_timing_prompt_processing(timing_prompt_processing);
double timing_token_generation = result.result_json.at("timings").value("predicted_ms", 0.0);
reply.set_timing_token_generation(timing_token_generation);
}
// Log Request Correlation Id
LOG_VERBOSE("correlation:", {
{ "id", data["correlation_id"] }
@ -2448,6 +2455,13 @@ public:
reply->set_prompt_tokens(tokens_evaluated);
reply->set_tokens(tokens_predicted);
reply->set_message(completion_text);
if (result.result_json.contains("timings")) {
double timing_prompt_processing = result.result_json.at("timings").value("prompt_ms", 0.0);
reply->set_timing_prompt_processing(timing_prompt_processing);
double timing_token_generation = result.result_json.at("timings").value("predicted_ms", 0.0);
reply->set_timing_token_generation(timing_token_generation);
}
}
else
{