package nos.inference
Get desktop application:
View/edit binary Protocol Buffers messages
Service definition
rpc Ping (, )
nos_service.proto:33Check health status of the inference server.
message
nos_service.proto:10Ping / healthcheck response
string status = 1
(e.g. "ok" or "not_ok")
rpc GetServiceInfo (, )
nos_service.proto:36Get service information (version, release date etc.)
message
nos_service.proto:15Service information repsonse
string version = 1
(e.g. "0.1.0")
string runtime = 2
(e.g. "cpu", "gpu", "local" etc)
rpc ListModels (, )
nos_service.proto:39List available models from Hugging Face Hub
rpc LoadModel (, )
nos_service.proto:42/ Pre-load a model for inference
rpc GetModelCatalog (, )
nos_service.proto:45Get model information from the deployment
rpc GetModelInfo (, )
nos_service.proto:46rpc Run (, )
nos_service.proto:49Run the inference request
rpc Stream (, stream )
nos_service.proto:52Stream the inference response
Register shared memory
Unregister shared memory
rpc UploadFile (stream , )
nos_service.proto:61Upload file
rpc DeleteFile (, )
nos_service.proto:64Delete file
message
nos_service.proto:21Register system shared memory request
Used as request type in: InferenceService.DeleteFile, InferenceService.LoadModel, InferenceService.RegisterSystemSharedMemory, InferenceService.Run, InferenceService.Stream, InferenceService.UnregisterSystemSharedMemory, InferenceService.UploadFile
bytes request_bytes = 1
message
nos_service.proto:26Register system shared memory response
Used as response type in: InferenceService.GetModelCatalog, InferenceService.GetModelInfo, InferenceService.ListModels, InferenceService.LoadModel, InferenceService.RegisterSystemSharedMemory, InferenceService.Run, InferenceService.Stream, InferenceService.UnregisterSystemSharedMemory, InferenceService.UploadFile
bytes response_bytes = 1