Skip to content

Allow configuration through argument list (in addition to string) #95

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
xenoscopic merged 1 commit into from
Jun 27, 2025
Merged

Allow configuration through argument list (in addition to string) #95

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions pkg/inference/backend.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,8 +30,8 @@ func (m BackendMode) String() string {
}

type BackendConfiguration struct {
ContextSize int64 `json:"context_size,omitempty"`
RawFlags []string `json:"flags,omitempty"`
ContextSize int64 `json:"context-size,omitempty"`
RuntimeFlags []string `json:"runtime-flags,omitempty"`
Comment on lines +33 to +34
Copy link
Contributor Author

@xenoscopic xenoscopic Jun 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't know if this JSON nomenclature is used anywhere (I couldn't find it), but I've tried to make it consistent before it becomes used anywhere.

}

// Backend is the interface implemented by inference engine backends. Backend
Expand Down
2 changes: 1 addition & 1 deletion pkg/inference/backends/llamacpp/llamacpp.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,7 +144,7 @@ func (l *llamaCpp) Run(ctx context.Context, socket, model string, mode inference
if config.ContextSize >= 0 {
args = append(args, "--ctx-size", strconv.Itoa(int(config.ContextSize)))
}
args = append(args, config.RawFlags...)
args = append(args, config.RuntimeFlags...)
}

l.log.Infof("llamaCppArgs: %v", args)
Expand Down
7 changes: 4 additions & 3 deletions pkg/inference/scheduling/api.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,7 +86,8 @@ type UnloadResponse struct {

// ConfigureRequest specifies per-model runtime configuration options.
type ConfigureRequest struct {
Model string `json:"model"`
ContextSize int64 `json:"context-size,omitempty"`
RawRuntimeFlags string `json:"raw-runtime-flags,omitempty"`
Model string `json:"model"`
ContextSize int64 `json:"context-size,omitempty"`
RuntimeFlags []string `json:"runtime-flags,omitempty"`
RawRuntimeFlags string `json:"raw-runtime-flags,omitempty"`
}
20 changes: 12 additions & 8 deletions pkg/inference/scheduling/scheduler.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -388,23 +388,27 @@ func (s *Scheduler) Configure(w http.ResponseWriter, r *http.Request) {
}

configureRequest := ConfigureRequest{
Model: "",
ContextSize: -1,
RawRuntimeFlags: "",
ContextSize: -1,
}
if err := json.Unmarshal(body, &configureRequest); err != nil {
http.Error(w, "invalid request", http.StatusBadRequest)
return
}
rawFlags, err := shellwords.Parse(configureRequest.RawRuntimeFlags)
if err != nil {
http.Error(w, "invalid request", http.StatusBadRequest)
return
var runtimeFlags []string
if len(configureRequest.RuntimeFlags) > 0 {
runtimeFlags = configureRequest.RuntimeFlags
} else {
rawFlags, err := shellwords.Parse(configureRequest.RawRuntimeFlags)
if err != nil {
http.Error(w, "invalid request", http.StatusBadRequest)
return
}
runtimeFlags = rawFlags
}

var runnerConfig inference.BackendConfiguration
runnerConfig.ContextSize = configureRequest.ContextSize
runnerConfig.RawFlags = rawFlags
runnerConfig.RuntimeFlags = runtimeFlags

if err := s.loader.setRunnerConfig(r.Context(), backend.Name(), configureRequest.Model, inference.BackendModeCompletion, runnerConfig); err != nil {
s.log.Warnf("Failed to configure %s runner for %s: %s", backend.Name(), configureRequest.Model, err)
Expand Down