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

set hf_repo/docker_repo as model alias when posible

This commit is contained in:
Xuan Son Nguyen 2025-11-26 15:57:20 +01:00
parent e40f35fb61
commit e2731c3767
5 changed files with 12 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
common/arg.cpp
View File

@ -218,6 +218,7 @@ static handle_model_result common_params_handle_model(
{ {
if (!model.docker_repo.empty()) { // Handle Docker URLs by resolving them to local paths if (!model.docker_repo.empty()) { // Handle Docker URLs by resolving them to local paths
model.path = common_docker_resolve_model(model.docker_repo); model.path = common_docker_resolve_model(model.docker_repo);
model.name = model.docker_repo; // set name for consistency
} else if (!model.hf_repo.empty()) { } else if (!model.hf_repo.empty()) {
// short-hand to avoid specifying --hf-file -> default it to --model // short-hand to avoid specifying --hf-file -> default it to --model
if (model.hf_file.empty()) { if (model.hf_file.empty()) {
@ -226,7 +227,8 @@ static handle_model_result common_params_handle_model(
if (auto_detected.repo.empty() || auto_detected.ggufFile.empty()) { if (auto_detected.repo.empty() || auto_detected.ggufFile.empty()) {
exit(1); // built without CURL, error message already printed exit(1); // built without CURL, error message already printed
} }
model.hf_repo = auto_detected.repo; model.name = model.hf_repo; // repo name with tag
model.hf_repo = auto_detected.repo; // repo name without tag
model.hf_file = auto_detected.ggufFile; model.hf_file = auto_detected.ggufFile;
if (!auto_detected.mmprojFile.empty()) { if (!auto_detected.mmprojFile.empty()) {
result.found_mmproj = true; result.found_mmproj = true;

1
common/common.h
View File

@ -203,6 +203,7 @@ struct common_params_model {
std::string hf_repo = ""; // HF repo // NOLINT std::string hf_repo = ""; // HF repo // NOLINT
std::string hf_file = ""; // HF file // NOLINT std::string hf_file = ""; // HF file // NOLINT
std::string docker_repo = ""; // Docker repo // NOLINT std::string docker_repo = ""; // Docker repo // NOLINT
std::string name = ""; // in format <user>/<model>[:<tag>] (tag is optional) // NOLINT
}; };
struct common_params_speculative { struct common_params_speculative {

4
common/download.h
View File

@ -14,8 +14,10 @@ struct common_cached_model_info {
std::string model; std::string model;
std::string tag; std::string tag;
size_t size = 0; // GGUF size in bytes size_t size = 0; // GGUF size in bytes
// return string representation like "user/model:tag"
// if tag is "latest", it will be omitted
std::string to_string() const { std::string to_string() const {
return user + "/" + model + ":" + tag; return user + "/" + model + (tag == "latest" ? "" : ":" + tag);
} }
}; };

1
tools/server/README.md
View File

@ -1445,7 +1445,6 @@ Listing all models in cache. The model metadata will also include a field to ind
```json ```json
{ {
"data": [{ "data": [{
"name": "ggml-org/gemma-3-4b-it-GGUF:Q4_K_M",
"id": "ggml-org/gemma-3-4b-it-GGUF:Q4_K_M", "id": "ggml-org/gemma-3-4b-it-GGUF:Q4_K_M",
"in_cache": true, "in_cache": true,
"path": "/Users/REDACTED/Library/Caches/llama.cpp/ggml-org_gemma-3-4b-it-GGUF_gemma-3-4b-it-Q4_K_M.gguf", "path": "/Users/REDACTED/Library/Caches/llama.cpp/ggml-org_gemma-3-4b-it-GGUF_gemma-3-4b-it-Q4_K_M.gguf",

6
tools/server/server.cpp
View File

@ -3353,7 +3353,6 @@ public:
} }
models_json.push_back(json { models_json.push_back(json {
{"id", meta.name}, {"id", meta.name},
{"name", meta.name},
{"object", "model"}, // for OAI-compat {"object", "model"}, // for OAI-compat
{"owned_by", "llamacpp"}, // for OAI-compat {"owned_by", "llamacpp"}, // for OAI-compat
{"created", t}, // for OAI-compat {"created", t}, // for OAI-compat
@ -3822,6 +3821,11 @@ int main(int argc, char ** argv, char ** envp) {
params.kv_unified = true; params.kv_unified = true;
} }
// for consistency between server router mode and single-model mode, we set the same model name as alias
if (params.model_alias.empty() && !params.model.name.empty()) {
params.model_alias = params.model.name;
}
common_init(); common_init();
// struct that contains llama context and inference // struct that contains llama context and inference