diff --git a/pkg/instance/models.go b/pkg/instance/models.go
new file mode 100644
index 0000000..f911f18
--- /dev/null
+++ b/pkg/instance/models.go
@@ -0,0 +1,141 @@
+package instance
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"llamactl/pkg/backends"
+	"net/http"
+	"time"
+)
+
+// Model represents a model available in a llama.cpp instance
+type Model struct {
+	ID      string      `json:"id"`
+	Object  string      `json:"object"`
+	OwnedBy string      `json:"owned_by"`
+	Created int64       `json:"created"`
+	InCache bool        `json:"in_cache"`
+	Path    string      `json:"path"`
+	Status  ModelStatus `json:"status"`
+}
+
+// ModelStatus represents the status of a model in an instance
+type ModelStatus struct {
+	Value string   `json:"value"` // "loaded" | "loading" | "unloaded"
+	Args  []string `json:"args"`
+}
+
+// IsLlamaCpp checks if this instance is a llama.cpp instance
+func (i *Instance) IsLlamaCpp() bool {
+	opts := i.GetOptions()
+	if opts == nil {
+		return false
+	}
+	return opts.BackendOptions.BackendType == backends.BackendTypeLlamaCpp
+}
+
+// GetModels fetches the models available in this llama.cpp instance
+func (i *Instance) GetModels() ([]Model, error) {
+	if !i.IsLlamaCpp() {
+		return nil, fmt.Errorf("instance %s is not a llama.cpp instance", i.Name)
+	}
+
+	if !i.IsRunning() {
+		return nil, fmt.Errorf("instance %s is not running", i.Name)
+	}
+
+	var result struct {
+		Data []Model `json:"data"`
+	}
+	if err := i.doRequest("GET", "/models", nil, &result, 10*time.Second); err != nil {
+		return nil, fmt.Errorf("failed to fetch models: %w", err)
+	}
+
+	return result.Data, nil
+}
+
+// LoadModel loads a model in this llama.cpp instance
+func (i *Instance) LoadModel(modelName string) error {
+	if !i.IsLlamaCpp() {
+		return fmt.Errorf("instance %s is not a llama.cpp instance", i.Name)
+	}
+
+	if !i.IsRunning() {
+		return fmt.Errorf("instance %s is not running", i.Name)
+	}
+
+	// Make the load request
+	reqBody := map[string]string{"model": modelName}
+	if err := i.doRequest("POST", "/models/load", reqBody, nil, 30*time.Second); err != nil {
+		return fmt.Errorf("failed to load model: %w", err)
+	}
+
+	return nil
+}
+
+// UnloadModel unloads a model from this llama.cpp instance
+func (i *Instance) UnloadModel(modelName string) error {
+	if !i.IsLlamaCpp() {
+		return fmt.Errorf("instance %s is not a llama.cpp instance", i.Name)
+	}
+
+	if !i.IsRunning() {
+		return fmt.Errorf("instance %s is not running", i.Name)
+	}
+
+	// Make the unload request
+	reqBody := map[string]string{"model": modelName}
+	if err := i.doRequest("POST", "/models/unload", reqBody, nil, 30*time.Second); err != nil {
+		return fmt.Errorf("failed to unload model: %w", err)
+	}
+
+	return nil
+}
+
+// doRequest makes an HTTP request to this instance's backend
+func (i *Instance) doRequest(method, path string, reqBody, respBody any, timeout time.Duration) error {
+	url := fmt.Sprintf("http://%s:%d%s", i.GetHost(), i.GetPort(), path)
+
+	var bodyReader io.Reader
+	if reqBody != nil {
+		bodyBytes, err := json.Marshal(reqBody)
+		if err != nil {
+			return fmt.Errorf("failed to marshal request body: %w", err)
+		}
+		bodyReader = bytes.NewReader(bodyBytes)
+	}
+
+	ctx, cancel := context.WithTimeout(context.Background(), timeout)
+	defer cancel()
+
+	req, err := http.NewRequestWithContext(ctx, method, url, bodyReader)
+	if err != nil {
+		return fmt.Errorf("failed to create request: %w", err)
+	}
+
+	if reqBody != nil {
+		req.Header.Set("Content-Type", "application/json")
+	}
+
+	resp, err := http.DefaultClient.Do(req)
+	if err != nil {
+		return err
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		bodyBytes, _ := io.ReadAll(resp.Body)
+		return fmt.Errorf("status %d: %s", resp.StatusCode, string(bodyBytes))
+	}
+
+	if respBody != nil {
+		if err := json.NewDecoder(resp.Body).Decode(respBody); err != nil {
+			return fmt.Errorf("failed to decode response: %w", err)
+		}
+	}
+
+	return nil
+}
diff --git a/pkg/manager/manager.go b/pkg/manager/manager.go
index 47e554b..73e747e 100644
--- a/pkg/manager/manager.go
+++ b/pkg/manager/manager.go
@@ -24,6 +24,8 @@ type InstanceManager interface {
 	EvictLRUInstance() error
 	RestartInstance(name string) (*instance.Instance, error)
 	GetInstanceLogs(name string, numLines int) (string, error)
+	ResolveInstance(modelName string) (string, error)
+	RefreshModelRegistry(inst *instance.Instance) error
 	Shutdown()
 }
 
@@ -34,6 +36,7 @@ type instanceManager struct {
 	db        database.InstanceStore
 	remote    *remoteManager
 	lifecycle *lifecycleManager
+	models    *modelRegistry
 
 	// Configuration
 	globalConfig *config.AppConfig
@@ -60,12 +63,16 @@ func New(globalConfig *config.AppConfig, db database.InstanceStore) InstanceMana
 	// Initialize remote manager
 	remote := newRemoteManager(globalConfig.Nodes, 30*time.Second)
 
+	// Initialize model registry
+	models := newModelRegistry()
+
 	// Create manager instance
 	im := &instanceManager{
 		registry:     registry,
 		ports:        ports,
 		db:           db,
 		remote:       remote,
+		models:       models,
 		globalConfig: globalConfig,
 	}
 
@@ -142,9 +149,27 @@ func (im *instanceManager) loadInstances() error {
 	// Auto-start instances that have auto-restart enabled
 	go im.autoStartInstances()
 
+	// Discover models from all running llama.cpp instances
+	go im.discoverAllModels()
+
 	return nil
 }
 
+// discoverAllModels discovers and registers models for all running llama.cpp instances
+func (im *instanceManager) discoverAllModels() {
+	instances := im.registry.listRunning()
+
+	for _, inst := range instances {
+		if !inst.IsLlamaCpp() {
+			continue
+		}
+
+		if err := im.RefreshModelRegistry(inst); err != nil {
+			log.Printf("Failed to discover models for instance %s: %v", inst.Name, err)
+		}
+	}
+}
+
 // loadInstance loads a single persisted instance and adds it to the registry
 func (im *instanceManager) loadInstance(persistedInst *instance.Instance) error {
 	name := persistedInst.Name
diff --git a/pkg/manager/model_registry.go b/pkg/manager/model_registry.go
new file mode 100644
index 0000000..0515c91
--- /dev/null
+++ b/pkg/manager/model_registry.go
@@ -0,0 +1,79 @@
+package manager
+
+import (
+	"fmt"
+	"llamactl/pkg/instance"
+	"sync"
+)
+
+// modelRegistry maintains a global mapping of model names to instance names
+// for llama.cpp instances. Model names must be globally unique across all instances.
+type modelRegistry struct {
+	mu              sync.RWMutex
+	modelToInstance map[string]string   // model name → instance name
+	instanceModels  map[string][]string // instance name → model names
+}
+
+// newModelRegistry creates a new model registry
+func newModelRegistry() *modelRegistry {
+	return &modelRegistry{
+		modelToInstance: make(map[string]string),
+		instanceModels:  make(map[string][]string),
+	}
+}
+
+// registerModels registers models from an instance to the registry.
+// Skips models that conflict with other instances and returns a list of conflicts.
+func (mr *modelRegistry) registerModels(instanceName string, models []instance.Model) []string {
+	mr.mu.Lock()
+	defer mr.mu.Unlock()
+
+	// Unregister any existing models for this instance first
+	mr.removeModels(instanceName)
+
+	// Register models, skipping conflicts
+	var modelNames []string
+	var conflicts []string
+
+	for _, model := range models {
+		// Check if this model conflicts with another instance
+		if existingInstance, exists := mr.modelToInstance[model.ID]; exists && existingInstance != instanceName {
+			conflicts = append(conflicts, fmt.Sprintf("%s (already in %s)", model.ID, existingInstance))
+			continue // Skip this model
+		}
+
+		// Register the model
+		mr.modelToInstance[model.ID] = instanceName
+		modelNames = append(modelNames, model.ID)
+	}
+
+	mr.instanceModels[instanceName] = modelNames
+
+	return conflicts
+}
+
+// unregisterModels removes all models for an instance
+func (mr *modelRegistry) unregisterModels(instanceName string) {
+	mr.mu.Lock()
+	defer mr.mu.Unlock()
+	mr.removeModels(instanceName)
+}
+
+// removeModels removes all models for an instance (caller must hold lock)
+func (mr *modelRegistry) removeModels(instanceName string) {
+	if models, exists := mr.instanceModels[instanceName]; exists {
+		for _, modelName := range models {
+			delete(mr.modelToInstance, modelName)
+		}
+		delete(mr.instanceModels, instanceName)
+	}
+}
+
+// getModelInstance returns the instance name that hosts the given model
+func (mr *modelRegistry) getModelInstance(modelName string) (string, bool) {
+	mr.mu.RLock()
+	defer mr.mu.RUnlock()
+
+	instanceName, exists := mr.modelToInstance[modelName]
+	return instanceName, exists
+}
diff --git a/pkg/manager/operations.go b/pkg/manager/operations.go
index 2cfbebf..6ff50f1 100644
--- a/pkg/manager/operations.go
+++ b/pkg/manager/operations.go
@@ -337,6 +337,9 @@ func (im *instanceManager) DeleteInstance(name string) error {
 	// Release port (use ReleaseByInstance for proper cleanup)
 	im.ports.releaseByInstance(name)
 
+	// Unregister models when instance is deleted
+	im.onInstanceStopped(name)
+
 	// Remove from registry
 	if err := im.registry.remove(name); err != nil {
 		return fmt.Errorf("failed to remove instance from registry: %w", err)
@@ -396,6 +399,9 @@ func (im *instanceManager) StartInstance(name string) (*instance.Instance, error
 		log.Printf("Warning: failed to persist instance %s: %v", name, err)
 	}
 
+	// Discover and register models for llama.cpp instances
+	go im.onInstanceStarted(name)
+
 	return inst, nil
 }
 
@@ -455,6 +461,9 @@ func (im *instanceManager) StopInstance(name string) (*instance.Instance, error)
 		log.Printf("Warning: failed to persist instance %s: %v", name, err)
 	}
 
+	// Unregister models when instance stops
+	im.onInstanceStopped(name)
+
 	return inst, nil
 }
 
@@ -535,3 +544,73 @@ func (im *instanceManager) setPortInOptions(options *instance.Options, port int)
 func (im *instanceManager) EvictLRUInstance() error {
 	return im.lifecycle.evictLRU()
 }
+
+// ResolveInstance resolves a model name to an instance name.
+// Precedence: instance name > model registry
+func (im *instanceManager) ResolveInstance(modelName string) (string, error) {
+	// Check if it's an instance name first
+	if _, err := im.GetInstance(modelName); err == nil {
+		return modelName, nil
+	}
+
+	// Check if it's a model name in the registry
+	if instanceName, exists := im.models.getModelInstance(modelName); exists {
+		return instanceName, nil
+	}
+
+	return "", fmt.Errorf("model or instance '%s' not found", modelName)
+}
+
+// RefreshModelRegistry refreshes the model registry for the given instance
+func (im *instanceManager) RefreshModelRegistry(inst *instance.Instance) error {
+	if !inst.IsRunning() {
+		return fmt.Errorf("instance %s is not running", inst.Name)
+	}
+
+	// Fetch models from instance and register them
+	models, err := inst.GetModels()
+	if err != nil {
+		return fmt.Errorf("failed to fetch models: %w", err)
+	}
+
+	// Register models, skipping conflicts
+	conflicts := im.models.registerModels(inst.Name, models)
+	if len(conflicts) > 0 {
+		log.Printf("Warning: Model name conflicts for instance %s (skipped): %v", inst.Name, conflicts)
+	}
+
+	// Check if instance name shadows any model names
+	if otherInstance, exists := im.models.getModelInstance(inst.Name); exists && otherInstance != inst.Name {
+		log.Printf("Warning: Instance name '%s' shadows model name from instance '%s'", inst.Name, otherInstance)
+	}
+
+	return nil
+}
+
+// onInstanceStarted is called when an instance successfully starts and becomes healthy
+func (im *instanceManager) onInstanceStarted(name string) {
+	inst, err := im.GetInstance(name)
+	if err != nil {
+		log.Printf("Failed to get instance %s for model discovery: %v", name, err)
+		return
+	}
+
+	// Only discover models for llama.cpp instances
+	if !inst.IsLlamaCpp() {
+		return
+	}
+
+	if err := inst.WaitForHealthy(30); err != nil {
+		log.Printf("Instance %s not healthy, skipping model discovery: %v", name, err)
+		return
+	}
+
+	if err := im.RefreshModelRegistry(inst); err != nil {
+		log.Printf("Failed to discover models for instance %s: %v", name, err)
+	}
+}
+
+// onInstanceStopped is called when an instance stops or is deleted
+func (im *instanceManager) onInstanceStopped(name string) {
+	im.models.unregisterModels(name)
+}
diff --git a/pkg/server/handlers_backends.go b/pkg/server/handlers_backends.go
index 065b24e..f912a8a 100644
--- a/pkg/server/handlers_backends.go
+++ b/pkg/server/handlers_backends.go
@@ -5,9 +5,12 @@ import (
 	"fmt"
 	"llamactl/pkg/backends"
 	"llamactl/pkg/instance"
+	"log"
 	"net/http"
 	"os/exec"
 	"strings"
+
+	"github.com/go-chi/chi/v5"
 )
 
 // ParseCommandRequest represents the request body for backend command parsing
@@ -306,3 +309,115 @@ func (h *Handler) LlamaServerVersionHandler() http.HandlerFunc {
 func (h *Handler) LlamaServerListDevicesHandler() http.HandlerFunc {
 	return h.executeLlamaServerCommand("--list-devices", "Failed to list devices")
 }
+
+// LlamaCppListModels godoc
+// @Summary List models in a llama.cpp instance
+// @Description Returns a list of models available in the specified llama.cpp instance
+// @Tags Llama.cpp
+// @Security ApiKeyAuth
+// @Produces json
+// @Param name path string true "Instance Name"
+// @Success 200 {object} map[string]any "Models list response"
+// @Failure 400 {string} string "Invalid instance"
+// @Failure 500 {string} string "Internal Server Error"
+// @Router /api/v1/llama-cpp/{name}/models [get]
+func (h *Handler) LlamaCppListModels() http.HandlerFunc {
+	return func(w http.ResponseWriter, r *http.Request) {
+		inst, err := h.getInstance(r)
+		if err != nil {
+			writeError(w, http.StatusBadRequest, "invalid_instance", err.Error())
+			return
+		}
+
+		models, err := inst.GetModels()
+		if err != nil {
+			writeError(w, http.StatusBadRequest, "get_models_failed", err.Error())
+			return
+		}
+
+		response := map[string]any{
+			"object": "list",
+			"data":   models,
+		}
+
+		writeJSON(w, http.StatusOK, response)
+	}
+}
+
+// LlamaCppLoadModel godoc
+// @Summary Load a model in a llama.cpp instance
+// @Description Loads the specified model in the given llama.cpp instance
+// @Tags Llama.cpp
+// @Security ApiKeyAuth
+// @Produces json
+// @Param name path string true "Instance Name"
+// @Param model path string true "Model Name"
+// @Success 200 {object} map[string]string "Success message"
+// @Failure 400 {string} string "Invalid request"
+// @Failure 500 {string} string "Internal Server Error"
+// @Router /api/v1/llama-cpp/{name}/models/{model}/load [post]
+func (h *Handler) LlamaCppLoadModel() http.HandlerFunc {
+	return func(w http.ResponseWriter, r *http.Request) {
+		inst, err := h.getInstance(r)
+		if err != nil {
+			writeError(w, http.StatusBadRequest, "invalid_instance", err.Error())
+			return
+		}
+
+		modelName := chi.URLParam(r, "model")
+
+		if err := inst.LoadModel(modelName); err != nil {
+			writeError(w, http.StatusBadRequest, "load_model_failed", err.Error())
+			return
+		}
+
+		// Refresh the model registry
+		if err := h.InstanceManager.RefreshModelRegistry(inst); err != nil {
+			log.Printf("Warning: failed to refresh model registry after load: %v", err)
+		}
+
+		writeJSON(w, http.StatusOK, map[string]string{
+			"status":  "success",
+			"message": fmt.Sprintf("Model %s loaded successfully", modelName),
+		})
+	}
+}
+
+// LlamaCppUnloadModel godoc
+// @Summary Unload a model in a llama.cpp instance
+// @Description Unloads the specified model in the given llama.cpp instance
+// @Tags Llama.cpp
+// @Security ApiKeyAuth
+// @Produces json
+// @Param name path string true "Instance Name"
+// @Param model path string true "Model Name"
+// @Success 200 {object} map[string]string "Success message"
+// @Failure 400 {string} string "Invalid request"
+// @Failure 500 {string} string "Internal Server Error"
+// @Router /api/v1/llama-cpp/{name}/models/{model}/unload [post]
+func (h *Handler) LlamaCppUnloadModel() http.HandlerFunc {
+	return func(w http.ResponseWriter, r *http.Request) {
+		inst, err := h.getInstance(r)
+		if err != nil {
+			writeError(w, http.StatusBadRequest, "invalid_instance", err.Error())
+			return
+		}
+
+		modelName := chi.URLParam(r, "model")
+
+		if err := inst.UnloadModel(modelName); err != nil {
+			writeError(w, http.StatusBadRequest, "unload_model_failed", err.Error())
+			return
+		}
+
+		// Refresh the model registry
+		if err := h.InstanceManager.RefreshModelRegistry(inst); err != nil {
+			log.Printf("Warning: failed to refresh model registry after unload: %v", err)
+		}
+
+		writeJSON(w, http.StatusOK, map[string]string{
+			"status":  "success",
+			"message": fmt.Sprintf("Model %s unloaded successfully", modelName),
+		})
+	}
+}
diff --git a/pkg/server/handlers_openai.go b/pkg/server/handlers_openai.go
index a7ad635..2240020 100644
--- a/pkg/server/handlers_openai.go
+++ b/pkg/server/handlers_openai.go
@@ -3,6 +3,7 @@ package server
 import (
 	"bytes"
 	"encoding/json"
+	"fmt"
 	"io"
 	"llamactl/pkg/instance"
 	"llamactl/pkg/validation"
@@ -40,14 +41,41 @@ func (h *Handler) OpenAIListInstances() http.HandlerFunc {
 			return
 		}
 
-		openaiInstances := make([]OpenAIInstance, len(instances))
-		for i, inst := range instances {
-			openaiInstances[i] = OpenAIInstance{
+		var openaiInstances []OpenAIInstance
+
+		// For each llama.cpp instance, try to fetch models and add them as separate entries
+		for _, inst := range instances {
+
+			if inst.IsLlamaCpp() && inst.IsRunning() {
+				// Try to fetch models from the instance
+				models, err := inst.GetModels()
+				if err != nil {
+					fmt.Printf("Failed to fetch models from instance %s: %v", inst.Name, err)
+					continue
+				}
+
+				for _, model := range models {
+					openaiInstances = append(openaiInstances, OpenAIInstance{
+						ID:      model.ID,
+						Object:  "model",
+						Created: model.Created,
+						OwnedBy: inst.Name,
+					})
+				}
+
+				if len(models) > 1 {
+					// Skip adding the instance name if multiple models are present
+					continue
+				}
+			}
+
+			// Add instance name as single entry (for non-llama.cpp or if model fetch failed)
+			openaiInstances = append(openaiInstances, OpenAIInstance{
 				ID:      inst.Name,
 				Object:  "model",
 				Created: inst.Created,
 				OwnedBy: "llamactl",
-			}
+			})
 		}
 
 		openaiResponse := OpenAIListInstancesResponse{
@@ -89,12 +117,19 @@ func (h *Handler) OpenAIProxy() http.HandlerFunc {
 
 		modelName, ok := requestBody["model"].(string)
 		if !ok || modelName == "" {
-			writeError(w, http.StatusBadRequest, "invalid_request", "Instance name is required")
+			writeError(w, http.StatusBadRequest, "invalid_request", "Model name is required")
+			return
+		}
+
+		// Resolve model name to instance name (checks instance names first, then model registry)
+		instanceName, err := h.InstanceManager.ResolveInstance(modelName)
+		if err != nil {
+			writeError(w, http.StatusBadRequest, "model_not_found", err.Error())
 			return
 		}
 
 		// Validate instance name at the entry point
-		validatedName, err := validation.ValidateInstanceName(modelName)
+		validatedName, err := validation.ValidateInstanceName(instanceName)
 		if err != nil {
 			writeError(w, http.StatusBadRequest, "invalid_instance_name", err.Error())
 			return
diff --git a/pkg/server/routes.go b/pkg/server/routes.go
index d8f8c35..81cca65 100644
--- a/pkg/server/routes.go
+++ b/pkg/server/routes.go
@@ -73,6 +73,13 @@ func SetupRouter(handler *Handler) *chi.Mux {
 			})
 		})
 
+		// Llama.cpp instance-specific endpoints
+		r.Route("/llama-cpp/{name}", func(r chi.Router) {
+			r.Get("/models", handler.LlamaCppListModels())
+			r.Post("/models/{model}/load", handler.LlamaCppLoadModel())
+			r.Post("/models/{model}/unload", handler.LlamaCppUnloadModel())
+		})
+
 		// Node management endpoints
 		r.Route("/nodes", func(r chi.Router) {
 			r.Get("/", handler.ListNodes()) // List all nodes
diff --git a/webui/src/components/InstanceCard.tsx b/webui/src/components/InstanceCard.tsx
index d889655..a79b116 100644
--- a/webui/src/components/InstanceCard.tsx
+++ b/webui/src/components/InstanceCard.tsx
@@ -2,13 +2,14 @@
 import { Button } from "@/components/ui/button";
 import { Card, CardContent, CardHeader, CardTitle } from "@/components/ui/card";
 import type { Instance } from "@/types/instance";
-import { Edit, FileText, Play, Square, Trash2, MoreHorizontal, Download } from "lucide-react";
+import { Edit, FileText, Play, Square, Trash2, MoreHorizontal, Download, Boxes } from "lucide-react";
 import LogsDialog from "@/components/LogDialog";
+import ModelsDialog from "@/components/ModelsDialog";
 import HealthBadge from "@/components/HealthBadge";
 import BackendBadge from "@/components/BackendBadge";
-import { useState } from "react";
+import { useState, useEffect } from "react";
 import { useInstanceHealth } from "@/hooks/useInstanceHealth";
-import { instancesApi } from "@/lib/api";
+import { instancesApi, llamaCppApi } from "@/lib/api";
 
 interface InstanceCardProps {
   instance: Instance;
@@ -26,9 +27,31 @@ function InstanceCard({
   editInstance,
 }: InstanceCardProps) {
   const [isLogsOpen, setIsLogsOpen] = useState(false);
+  const [isModelsOpen, setIsModelsOpen] = useState(false);
   const [showAllActions, setShowAllActions] = useState(false);
+  const [modelCount, setModelCount] = useState(0);
   const health = useInstanceHealth(instance.name, instance.status);
 
+  const running = instance.status === "running";
+  const isLlamaCpp = instance.options?.backend_type === "llama_cpp";
+
+  // Fetch model count for llama.cpp instances
+  useEffect(() => {
+    if (!isLlamaCpp || !running) {
+      setModelCount(0);
+      return;
+    }
+
+    void (async () => {
+      try {
+        const models = await llamaCppApi.getModels(instance.name);
+        setModelCount(models.length);
+      } catch {
+        setModelCount(0);
+      }
+    })();
+  }, [instance.name, isLlamaCpp, running]);
+
   const handleStart = () => {
     startInstance(instance.name);
   };
@@ -53,6 +76,10 @@ function InstanceCard({
     setIsLogsOpen(true);
   };
 
+  const handleModels = () => {
+    setIsModelsOpen(true);
+  };
+
   const handleExport = () => {
     void (async () => {
       try {
@@ -83,8 +110,6 @@ function InstanceCard({
     })();
   };
 
-  const running = instance.status === "running";
-
   return (
     <>
       <Card className="hover:shadow-md transition-shadow">
@@ -162,6 +187,20 @@ function InstanceCard({
                 Logs
               </Button>
 
+              {isLlamaCpp && modelCount > 1 && (
+                <Button
+                  size="sm"
+                  variant="outline"
+                  onClick={handleModels}
+                  title="Manage models"
+                  data-testid="manage-models-button"
+                  className="flex-1"
+                >
+                  <Boxes className="h-4 w-4 mr-1" />
+                  Models ({modelCount})
+                </Button>
+              )}
+
               <Button
                 size="sm"
                 variant="outline"
@@ -195,6 +234,13 @@ function InstanceCard({
         instanceName={instance.name}
         isRunning={running}
       />
+
+      <ModelsDialog
+        open={isModelsOpen}
+        onOpenChange={setIsModelsOpen}
+        instanceName={instance.name}
+        isRunning={running}
+      />
     </>
   );
 }
diff --git a/webui/src/components/ModelsDialog.tsx b/webui/src/components/ModelsDialog.tsx
new file mode 100644
index 0000000..d5ba733
--- /dev/null
+++ b/webui/src/components/ModelsDialog.tsx
@@ -0,0 +1,287 @@
+import React, { useState, useEffect } from 'react'
+import { Button } from '@/components/ui/button'
+import {
+  Dialog,
+  DialogContent,
+  DialogDescription,
+  DialogHeader,
+  DialogTitle,
+} from '@/components/ui/dialog'
+import {
+  Table,
+  TableBody,
+  TableCell,
+  TableHead,
+  TableHeader,
+  TableRow,
+} from '@/components/ui/table'
+import { Badge } from '@/components/ui/badge'
+import { llamaCppApi } from '@/lib/api'
+import { RefreshCw, Loader2, AlertCircle } from 'lucide-react'
+
+interface ModelsDialogProps {
+  open: boolean
+  onOpenChange: (open: boolean) => void
+  instanceName: string
+  isRunning: boolean
+}
+
+interface Model {
+  id: string
+  object: string
+  owned_by: string
+  created: number
+  in_cache: boolean
+  path: string
+  status: {
+    value: string // "loaded" | "loading" | "unloaded"
+    args: string[]
+  }
+}
+
+const StatusIcon: React.FC<{ status: string }> = ({ status }) => {
+  switch (status) {
+    case 'loaded':
+      return (
+        <div className="h-2 w-2 rounded-full bg-green-500" />
+      )
+    case 'loading':
+      return (
+        <Loader2
+          className="h-3 w-3 animate-spin text-yellow-500"
+        />
+      )
+    case 'unloaded':
+      return (
+        <div className="h-2 w-2 rounded-full bg-gray-400" />
+      )
+    default:
+      return null
+  }
+}
+
+const ModelsDialog: React.FC<ModelsDialogProps> = ({
+  open,
+  onOpenChange,
+  instanceName,
+  isRunning,
+}) => {
+  const [models, setModels] = useState<Model[]>([])
+  const [loading, setLoading] = useState(false)
+  const [error, setError] = useState<string | null>(null)
+  const [loadingModels, setLoadingModels] = useState<Set<string>>(new Set())
+
+  // Fetch models function
+  const fetchModels = React.useCallback(async () => {
+    if (!instanceName || !isRunning) return
+
+    setLoading(true)
+    setError(null)
+
+    try {
+      const response = await llamaCppApi.getModels(instanceName)
+      setModels(response)
+    } catch (err) {
+      setError(err instanceof Error ? err.message : 'Failed to fetch models')
+    } finally {
+      setLoading(false)
+    }
+  }, [instanceName, isRunning])
+
+  // Poll for models while dialog is open
+  useEffect(() => {
+    if (!open || !isRunning) return
+
+    // Initial fetch
+    void fetchModels()
+
+    // Poll every 2 seconds
+    const interval = setInterval(() => {
+      void fetchModels()
+    }, 2000)
+
+    return () => clearInterval(interval)
+  }, [open, isRunning, fetchModels])
+
+  // Load model
+  const loadModel = async (modelName: string) => {
+    setLoadingModels((prev) => new Set(prev).add(modelName))
+    setError(null)
+
+    try {
+      await llamaCppApi.loadModel(instanceName, modelName)
+      // Polling will pick up the change
+    } catch (err) {
+      setError(err instanceof Error ? err.message : 'Failed to load model')
+    } finally {
+      setLoadingModels((prev) => {
+        const newSet = new Set(prev)
+        newSet.delete(modelName)
+        return newSet
+      })
+    }
+  }
+
+  // Unload model
+  const unloadModel = async (modelName: string) => {
+    setLoadingModels((prev) => new Set(prev).add(modelName))
+    setError(null)
+
+    try {
+      await llamaCppApi.unloadModel(instanceName, modelName)
+      // Polling will pick up the change
+    } catch (err) {
+      setError(err instanceof Error ? err.message : 'Failed to unload model')
+    } finally {
+      setLoadingModels((prev) => {
+        const newSet = new Set(prev)
+        newSet.delete(modelName)
+        return newSet
+      })
+    }
+  }
+
+  return (
+    <Dialog open={open} onOpenChange={onOpenChange}>
+      <DialogContent className="sm:max-w-4xl max-w-[calc(100%-2rem)] max-h-[80vh] flex flex-col">
+        <DialogHeader>
+          <div className="flex items-center justify-between">
+            <div>
+              <DialogTitle className="flex items-center gap-2">
+                Models: {instanceName}
+                <Badge variant={isRunning ? 'default' : 'secondary'}>
+                  {isRunning ? 'Running' : 'Stopped'}
+                </Badge>
+              </DialogTitle>
+              <DialogDescription>
+                Manage models in this llama.cpp instance
+              </DialogDescription>
+            </div>
+
+            <Button
+              variant="outline"
+              size="sm"
+              onClick={() => void fetchModels()}
+              disabled={loading || !isRunning}
+            >
+              {loading ? (
+                <Loader2 className="h-4 w-4 animate-spin" />
+              ) : (
+                <RefreshCw className="h-4 w-4" />
+              )}
+            </Button>
+          </div>
+        </DialogHeader>
+
+        {/* Error Display */}
+        {error && (
+          <div className="flex items-center gap-2 p-3 bg-destructive/10 border border-destructive/20 rounded-lg">
+            <AlertCircle className="h-4 w-4 text-destructive" />
+            <span className="text-sm text-destructive">{error}</span>
+          </div>
+        )}
+
+        {/* Models Table */}
+        <div className="flex-1 flex flex-col min-h-0 overflow-auto">
+          {!isRunning ? (
+            <div className="flex items-center justify-center h-full text-muted-foreground">
+              Instance is not running
+            </div>
+          ) : loading && models.length === 0 ? (
+            <div className="flex items-center justify-center h-full">
+              <Loader2 className="h-6 w-6 animate-spin text-muted-foreground" />
+              <span className="ml-2 text-muted-foreground">
+                Loading models...
+              </span>
+            </div>
+          ) : models.length === 0 ? (
+            <div className="flex items-center justify-center h-full text-muted-foreground">
+              No models found
+            </div>
+          ) : (
+            <Table>
+              <TableHeader>
+                <TableRow>
+                  <TableHead>Model</TableHead>
+                  <TableHead>Status</TableHead>
+                  <TableHead className="text-right">Actions</TableHead>
+                </TableRow>
+              </TableHeader>
+              <TableBody>
+                {models.map((model) => {
+                  const isLoading = loadingModels.has(model.id)
+                  const isModelLoading = model.status.value === 'loading'
+
+                  return (
+                    <TableRow key={model.id}>
+                      <TableCell className="font-mono text-sm">
+                        {model.id}
+                      </TableCell>
+                      <TableCell>
+                        <div className="flex items-center gap-2">
+                          <StatusIcon status={model.status.value} />
+                          <span className="text-sm capitalize">
+                            {model.status.value}
+                          </span>
+                        </div>
+                      </TableCell>
+                      <TableCell className="text-right">
+                        {model.status.value === 'loaded' ? (
+                          <Button
+                            size="sm"
+                            variant="outline"
+                            onClick={() => unloadModel(model.id)}
+                            disabled={!isRunning || isLoading || isModelLoading}
+                          >
+                            {isLoading ? (
+                              <>
+                                <Loader2 className="h-3 w-3 animate-spin mr-1" />
+                                Unloading...
+                              </>
+                            ) : (
+                              'Unload'
+                            )}
+                          </Button>
+                        ) : model.status.value === 'unloaded' ? (
+                          <Button
+                            size="sm"
+                            variant="default"
+                            onClick={() => loadModel(model.id)}
+                            disabled={!isRunning || isLoading || isModelLoading}
+                          >
+                            {isLoading ? (
+                              <>
+                                <Loader2 className="h-3 w-3 animate-spin mr-1" />
+                                Loading...
+                              </>
+                            ) : (
+                              'Load'
+                            )}
+                          </Button>
+                        ) : (
+                          <Button size="sm" variant="ghost" disabled>
+                            Loading...
+                          </Button>
+                        )}
+                      </TableCell>
+                    </TableRow>
+                  )
+                })}
+              </TableBody>
+            </Table>
+          )}
+        </div>
+
+        {/* Auto-refresh indicator */}
+        {isRunning && (
+          <div className="flex items-center gap-2 text-sm text-muted-foreground">
+            <div className="w-2 h-2 bg-green-500 rounded-full animate-pulse"></div>
+            Auto-refreshing every 2 seconds
+          </div>
+        )}
+      </DialogContent>
+    </Dialog>
+  )
+}
+
+export default ModelsDialog
diff --git a/webui/src/components/ui/table.tsx b/webui/src/components/ui/table.tsx
new file mode 100644
index 0000000..7f3502f
--- /dev/null
+++ b/webui/src/components/ui/table.tsx
@@ -0,0 +1,117 @@
+import * as React from "react"
+
+import { cn } from "@/lib/utils"
+
+const Table = React.forwardRef<
+  HTMLTableElement,
+  React.HTMLAttributes<HTMLTableElement>
+>(({ className, ...props }, ref) => (
+  <div className="relative w-full overflow-auto">
+    <table
+      ref={ref}
+      className={cn("w-full caption-bottom text-sm", className)}
+      {...props}
+    />
+  </div>
+))
+Table.displayName = "Table"
+
+const TableHeader = React.forwardRef<
+  HTMLTableSectionElement,
+  React.HTMLAttributes<HTMLTableSectionElement>
+>(({ className, ...props }, ref) => (
+  <thead ref={ref} className={cn("[&_tr]:border-b", className)} {...props} />
+))
+TableHeader.displayName = "TableHeader"
+
+const TableBody = React.forwardRef<
+  HTMLTableSectionElement,
+  React.HTMLAttributes<HTMLTableSectionElement>
+>(({ className, ...props }, ref) => (
+  <tbody
+    ref={ref}
+    className={cn("[&_tr:last-child]:border-0", className)}
+    {...props}
+  />
+))
+TableBody.displayName = "TableBody"
+
+const TableFooter = React.forwardRef<
+  HTMLTableSectionElement,
+  React.HTMLAttributes<HTMLTableSectionElement>
+>(({ className, ...props }, ref) => (
+  <tfoot
+    ref={ref}
+    className={cn(
+      "border-t bg-muted/50 font-medium [&>tr]:last:border-b-0",
+      className
+    )}
+    {...props}
+  />
+))
+TableFooter.displayName = "TableFooter"
+
+const TableRow = React.forwardRef<
+  HTMLTableRowElement,
+  React.HTMLAttributes<HTMLTableRowElement>
+>(({ className, ...props }, ref) => (
+  <tr
+    ref={ref}
+    className={cn(
+      "border-b transition-colors hover:bg-muted/50 data-[state=selected]:bg-muted",
+      className
+    )}
+    {...props}
+  />
+))
+TableRow.displayName = "TableRow"
+
+const TableHead = React.forwardRef<
+  HTMLTableCellElement,
+  React.ThHTMLAttributes<HTMLTableCellElement>
+>(({ className, ...props }, ref) => (
+  <th
+    ref={ref}
+    className={cn(
+      "h-12 px-4 text-left align-middle font-medium text-muted-foreground [&:has([role=checkbox])]:pr-0",
+      className
+    )}
+    {...props}
+  />
+))
+TableHead.displayName = "TableHead"
+
+const TableCell = React.forwardRef<
+  HTMLTableCellElement,
+  React.TdHTMLAttributes<HTMLTableCellElement>
+>(({ className, ...props }, ref) => (
+  <td
+    ref={ref}
+    className={cn("p-4 align-middle [&:has([role=checkbox])]:pr-0", className)}
+    {...props}
+  />
+))
+TableCell.displayName = "TableCell"
+
+const TableCaption = React.forwardRef<
+  HTMLTableCaptionElement,
+  React.HTMLAttributes<HTMLTableCaptionElement>
+>(({ className, ...props }, ref) => (
+  <caption
+    ref={ref}
+    className={cn("mt-4 text-sm text-muted-foreground", className)}
+    {...props}
+  />
+))
+TableCaption.displayName = "TableCaption"
+
+export {
+  Table,
+  TableHeader,
+  TableBody,
+  TableFooter,
+  TableHead,
+  TableRow,
+  TableCell,
+  TableCaption,
+}
diff --git a/webui/src/lib/api.ts b/webui/src/lib/api.ts
index ddcf384..aeeb100 100644
--- a/webui/src/lib/api.ts
+++ b/webui/src/lib/api.ts
@@ -205,3 +205,51 @@ export const apiKeysApi = {
   getPermissions: (id: number) =>
     apiCall<KeyPermissionResponse[]>(`/auth/keys/${id}/permissions`),
 };
+
+// Llama.cpp model management types
+export interface Model {
+  id: string;
+  object: string;
+  owned_by: string;
+  created: number;
+  in_cache: boolean;
+  path: string;
+  status: {
+    value: string; // "loaded" | "loading" | "unloaded"
+    args: string[];
+  };
+}
+
+export interface ModelsListResponse {
+  object: string;
+  data: Model[];
+}
+
+// Llama.cpp model management API functions
+export const llamaCppApi = {
+  // GET /llama-cpp/{name}/models
+  getModels: async (instanceName: string): Promise<Model[]> => {
+    const response = await apiCall<ModelsListResponse>(
+      `/llama-cpp/${encodeURIComponent(instanceName)}/models`
+    );
+    return response.data;
+  },
+
+  // POST /llama-cpp/{name}/models/{model}/load
+  loadModel: (instanceName: string, modelName: string) =>
+    apiCall<{ status: string; message: string }>(
+      `/llama-cpp/${encodeURIComponent(instanceName)}/models/${encodeURIComponent(modelName)}/load`,
+      {
+        method: "POST",
+      }
+    ),
+
+  // POST /llama-cpp/{name}/models/{model}/unload
+  unloadModel: (instanceName: string, modelName: string) =>
+    apiCall<{ status: string; message: string }>(
+      `/llama-cpp/${encodeURIComponent(instanceName)}/models/${encodeURIComponent(modelName)}/unload`,
+      {
+        method: "POST",
+      }
+    ),
+};