happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

oai-compat /models endpoint

This commit is contained in:
Xuan Son Nguyen 2025-11-23 17:25:24 +01:00
parent f95f9c5128
commit 2e355c7f8e
1 changed files with 20 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
tools/server/server.cpp
View File

@ -5142,6 +5142,11 @@ public:
server_http_context::handler_t proxy_get = [this](const server_http_req & req) { server_http_context::handler_t proxy_get = [this](const server_http_req & req) {
std::string method = "GET"; std::string method = "GET";
std::string name = req.get_param("model"); std::string name = req.get_param("model");
if (name.empty()) {
auto res = std::make_unique<server_res_generator>(ctx_server);
res->error(format_error_response("model name is missing from the request", ERROR_TYPE_INVALID_REQUEST));
return std::unique_ptr<server_http_res>(std::move(res));
}
models->ensure_model_loaded(name); models->ensure_model_loaded(name);
return models->proxy_request(req, method, name, false); return models->proxy_request(req, method, name, false);
}; };
@ -5150,6 +5155,11 @@ public:
std::string method = "POST"; std::string method = "POST";
json body = json::parse(req.body); json body = json::parse(req.body);
std::string name = json_value(body, "model", std::string()); std::string name = json_value(body, "model", std::string());
if (name.empty()) {
auto res = std::make_unique<server_res_generator>(ctx_server);
res->error(format_error_response("model name is missing from the request", ERROR_TYPE_INVALID_REQUEST));
return std::unique_ptr<server_http_res>(std::move(res));
}
models->ensure_model_loaded(name); models->ensure_model_loaded(name);
return models->proxy_request(req, method, name, true); // update last usage for POST request only return models->proxy_request(req, method, name, true); // update last usage for POST request only
}; };
@ -5189,6 +5199,7 @@ public:
auto res = std::make_unique<server_res_generator>(ctx_server); auto res = std::make_unique<server_res_generator>(ctx_server);
json models_json = json::array(); json models_json = json::array();
auto all_models = models->get_all_meta(); auto all_models = models->get_all_meta();
std::time_t t = std::time(0);
for (const auto & model : all_models) { for (const auto & model : all_models) {
json status { json status {
{"value", server_model_status_to_string(model.status)}, {"value", server_model_status_to_string(model.status)},
@ -5198,15 +5209,21 @@ public:
status["exit_code"] = model.exit_code; status["exit_code"] = model.exit_code;
} }
models_json.push_back(json { models_json.push_back(json {
{"name", model.name},
{"id", model.name}, {"id", model.name},
{"name", model.name},
{"object", "model"}, // for OAI-compat
{"owned_by", "llamacpp"}, // for OAI-compat
{"created", t}, // for OAI-compat
{"in_cache", model.in_cache}, {"in_cache", model.in_cache},
{"path", model.path}, {"path", model.path},
{"status", status}, {"status", status},
// TODO: other fields... // TODO: add other fields, may require reading GGUF metadata
}); });
} }
res->ok({{"data", models_json}}); res->ok({
{"data", models_json},
{"object", "list"},
});
return res; return res;
}; };