Skip to content

[pull] master from mudler:master #856

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
pull merged 2 commits into from
Jan 3, 2026
Merged

[pull] master from mudler:master #856

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5f6c941
fix(llama.cpp/mmproj): fix loading mmproj in nested sub-dirs differen…
mudler Jan 2, 2026
641606a
chore: :arrow_up: Update ggml-org/llama.cpp to `706e3f93a60109a40f122…
localai-bot Jan 2, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion backend/cpp/llama-cpp/Makefile
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@

LLAMA_VERSION?=ced765be44ce173c374f295b3c6f4175f8fd109b
LLAMA_VERSION?=706e3f93a60109a40f1224eaf4af0d59caa7c3ae
LLAMA_REPO?=https://github.com/ggerganov/llama.cpp

CMAKE_ARGS?=
Expand Down
4 changes: 1 addition & 3 deletions backend/cpp/llama-cpp/grpc-server.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -358,9 +358,7 @@ static void params_parse(server_context& /*ctx_server*/, const backend::ModelOpt

params.model.path = request->modelfile();
if (!request->mmproj().empty()) {
// get the directory of modelfile
std::string model_dir = params.model.path.substr(0, params.model.path.find_last_of("/\\"));
params.mmproj.path = model_dir + "/"+ request->mmproj();
params.mmproj.path = request->mmproj();
}
// params.model_alias ??
params.model_alias = request->modelfile();
Expand Down
13 changes: 9 additions & 4 deletions core/backend/options.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ func ModelOptions(c config.ModelConfig, so *config.ApplicationConfig, opts ...mo

c.Threads = &threads

grpcOpts := grpcModelOpts(c)
grpcOpts := grpcModelOpts(c, so.SystemState.Model.ModelsPath)
defOpts = append(defOpts, model.WithLoadGRPCLoadModelOpts(grpcOpts))

if so.ParallelBackendRequests {
Expand Down Expand Up @@ -72,7 +72,7 @@ func getSeed(c config.ModelConfig) int32 {
return seed
}

func grpcModelOpts(c config.ModelConfig) *pb.ModelOptions {
func grpcModelOpts(c config.ModelConfig, modelPath string) *pb.ModelOptions {
b := 512
if c.Batch != 0 {
b = c.Batch
Expand Down Expand Up @@ -131,7 +131,7 @@ func grpcModelOpts(c config.ModelConfig) *pb.ModelOptions {
})
}

return &pb.ModelOptions{
opts := &pb.ModelOptions{
CUDA: c.CUDA || c.Diffusers.CUDA,
SchedulerType: c.Diffusers.SchedulerType,
GrammarTriggers: triggers,
Expand Down Expand Up @@ -170,7 +170,6 @@ func grpcModelOpts(c config.ModelConfig) *pb.ModelOptions {
LimitImagePerPrompt: int32(c.LimitMMPerPrompt.LimitImagePerPrompt),
LimitVideoPerPrompt: int32(c.LimitMMPerPrompt.LimitVideoPerPrompt),
LimitAudioPerPrompt: int32(c.LimitMMPerPrompt.LimitAudioPerPrompt),
MMProj: c.MMProj,
FlashAttention: flashAttention,
CacheTypeKey: c.CacheTypeK,
CacheTypeValue: c.CacheTypeV,
Expand Down Expand Up @@ -198,6 +197,12 @@ func grpcModelOpts(c config.ModelConfig) *pb.ModelOptions {
// RWKV
Tokenizer: c.Tokenizer,
}

if c.MMProj != "" {
opts.MMProj = filepath.Join(modelPath, c.MMProj)
}

return opts
}

func gRPCPredictOpts(c config.ModelConfig, modelPath string) *pb.PredictOptions {
Expand Down
Loading