add model/load endpoint

Signed-off-by: Emily Casey <emily.casey@docker.com>

Author: ekcasey

go.mod

@@ -5,8 +5,10 @@ go 1.23.7
 require (
 	github.com/containerd/containerd/v2 v2.0.4
 	github.com/containerd/platforms v1.0.0-rc.1
-	github.com/docker/model-distribution v0.0.0-20250710123110-a633223e127e
+	github.com/docker/model-distribution v0.0.0-20250724035854-a9454ee7284c
+	github.com/elastic/go-sysinfo v1.15.3
 	github.com/google/go-containerregistry v0.20.3
+	github.com/gpustack/gguf-parser-go v0.14.1
 	github.com/jaypipes/ghw v0.16.0
 	github.com/mattn/go-shellwords v1.0.12
 	github.com/opencontainers/go-digest v1.0.0
@@ -29,13 +31,11 @@ require (
 	github.com/docker/cli v27.5.0+incompatible // indirect
 	github.com/docker/distribution v2.8.3+incompatible // indirect
 	github.com/docker/docker-credential-helpers v0.8.2 // indirect
-	github.com/elastic/go-sysinfo v1.15.3 // indirect
 	github.com/elastic/go-windows v1.0.2 // indirect
 	github.com/felixge/httpsnoop v1.0.4 // indirect
 	github.com/go-logr/logr v1.4.2 // indirect
 	github.com/go-logr/stdr v1.2.2 // indirect
 	github.com/go-ole/go-ole v1.2.6 // indirect
-	github.com/gpustack/gguf-parser-go v0.14.1 // indirect
 	github.com/henvic/httpretty v0.1.4 // indirect
 	github.com/jaypipes/pcidb v1.0.1 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect

go.sum

@@ -38,8 +38,8 @@ github.com/docker/distribution v2.8.3+incompatible h1:AtKxIZ36LoNK51+Z6RpzLpddBi
 github.com/docker/distribution v2.8.3+incompatible/go.mod h1:J2gT2udsDAN96Uj4KfcMRqY0/ypR+oyYUYmja8H+y+w=
 github.com/docker/docker-credential-helpers v0.8.2 h1:bX3YxiGzFP5sOXWc3bTPEXdEaZSeVMrFgOr3T+zrFAo=
 github.com/docker/docker-credential-helpers v0.8.2/go.mod h1:P3ci7E3lwkZg6XiHdRKft1KckHiO9a2rNtyFbZ/ry9M=
-github.com/docker/model-distribution v0.0.0-20250710123110-a633223e127e h1:qBkjP4A20f3RXvtstitIPiStQ4p+bK8xcjosrXLBQZ0=
-github.com/docker/model-distribution v0.0.0-20250710123110-a633223e127e/go.mod h1:dThpO9JoG5Px3i+rTluAeZcqLGw8C0qepuEL4gL2o/c=
+github.com/docker/model-distribution v0.0.0-20250724035854-a9454ee7284c h1:hkMon8WnL995BHacRCEsbtPEO2mYsUmdsohJva15QDU=
+github.com/docker/model-distribution v0.0.0-20250724035854-a9454ee7284c/go.mod h1:dThpO9JoG5Px3i+rTluAeZcqLGw8C0qepuEL4gL2o/c=
 github.com/elastic/go-sysinfo v1.15.3 h1:W+RnmhKFkqPTCRoFq2VCTmsT4p/fwpo+3gKNQsn1XU0=
 github.com/elastic/go-sysinfo v1.15.3/go.mod h1:K/cNrqYTDrSoMh2oDkYEMS2+a72GRxMvNP+GC+vRIlo=
 github.com/elastic/go-windows v1.0.2 h1:yoLLsAsV5cfg9FLhZ9EXZ2n2sQFKeDYrHenkcivY4vI=

pkg/inference/models/manager.go

@@ -15,10 +15,11 @@ import (
 	"github.com/docker/model-distribution/distribution"
 	"github.com/docker/model-distribution/registry"
 	"github.com/docker/model-distribution/types"
+	"github.com/sirupsen/logrus"
+
 	"github.com/docker/model-runner/pkg/diskusage"
 	"github.com/docker/model-runner/pkg/inference"
 	"github.com/docker/model-runner/pkg/logging"
-	"github.com/sirupsen/logrus"
 )
 
 const (
@@ -120,6 +121,7 @@ func (m *Manager) RebuildRoutes(allowedOrigins []string) {
 func (m *Manager) routeHandlers(allowedOrigins []string) map[string]http.HandlerFunc {
 	handlers := map[string]http.HandlerFunc{
 		"POST " + inference.ModelsPrefix + "/create":                          m.handleCreateModel,
+		"POST " + inference.ModelsPrefix + "/load":                            m.handleLoadModel,
 		"GET " + inference.ModelsPrefix:                                       m.handleGetModels,
 		"GET " + inference.ModelsPrefix + "/{name...}":                        m.handleGetModel,
 		"DELETE " + inference.ModelsPrefix + "/{name...}":                     m.handleDeleteModel,
@@ -187,6 +189,20 @@ func (m *Manager) handleCreateModel(w http.ResponseWriter, r *http.Request) {
 	}
 }
 
+// handleLoadModel handles POST <inference-prefix>/models/load requests.
+func (m *Manager) handleLoadModel(w http.ResponseWriter, r *http.Request) {
+	if m.distributionClient == nil {
+		http.Error(w, "model distribution service unavailable", http.StatusServiceUnavailable)
+		return
+	}
+
+	if _, err := m.distributionClient.LoadModel(r.Body, w); err != nil {
+		http.Error(w, err.Error(), http.StatusInternalServerError)
+		return
+	}
+	return
+}
+
 // handleGetModels handles GET <inference-prefix>/models requests.
 func (m *Manager) handleGetModels(w http.ResponseWriter, r *http.Request) {
 	if m.distributionClient == nil {