Add LightVAE support

frontend/src/components/SettingsPanel.tsx

@@ -34,6 +34,7 @@ import type {
   SettingsState,
   InputMode,
   PipelineInfo,
+  VaeType,
 } from "../types";
 import { LoRAManager } from "./LoRAManager";
 
@@ -87,6 +88,11 @@ interface SettingsPanelProps {
   onVaceEnabledChange?: (enabled: boolean) => void;
   vaceContextScale?: number;
   onVaceContextScaleChange?: (scale: number) => void;
+  // VAE type selection
+  vaeType?: VaeType;
+  onVaeTypeChange?: (vaeType: VaeType) => void;
+  // Available VAE types from backend registry
+  vaeTypes?: string[];
 }
 
 export function SettingsPanel({
@@ -126,6 +132,9 @@ export function SettingsPanel({
   onVaceEnabledChange,
   vaceContextScale = 1.0,
   onVaceContextScaleChange,
+  vaeType = "wan",
+  onVaeTypeChange,
+  vaeTypes = ["wan"],
 }: SettingsPanelProps) {
   // Local slider state management hooks
   const noiseScaleSlider = useLocalSliderValue(noiseScale, onNoiseScaleChange);
@@ -389,6 +398,37 @@ export function SettingsPanel({
           </div>
         )}
 
+        {/* VAE Type Selection */}
+        {vaeTypes && vaeTypes.length > 0 && (
+          <div className="space-y-2">
+            <div className="flex items-center justify-between gap-2">
+              <LabelWithTooltip
+                label={PARAMETER_METADATA.vaeType.label}
+                tooltip={PARAMETER_METADATA.vaeType.tooltip}
+                className="text-sm text-foreground"
+              />
+              <Select
+                value={vaeType}
+                onValueChange={value => {
+                  onVaeTypeChange?.(value as VaeType);
+                }}
+                disabled={isStreaming}
+              >
+                <SelectTrigger className="w-[140px] h-7">
+                  <SelectValue />
+                </SelectTrigger>
+                <SelectContent>
+                  {vaeTypes.map(type => (
+                    <SelectItem key={type} value={type}>
+                      {type}
+                    </SelectItem>
+                  ))}
+                </SelectContent>
+              </Select>
+            </div>
+          </div>
+        )}
+
         {currentPipeline?.supportsLoRA && (
           <div className="space-y-4">
             <LoRAManager

frontend/src/data/parameterMetadata.ts

@@ -82,4 +82,9 @@ export const PARAMETER_METADATA: Record<string, ParameterMetadata> = {
     tooltip:
       "The configuration of the sender that will send video to Spout-compatible apps like TouchDesigner, Resolume, OBS.",
   },
+  vaeType: {
+    label: "VAE:",
+    tooltip:
+      "VAE type to use for encoding/decoding. 'wan' is the full VAE with best quality. 'lightvae' is 75% pruned for faster performance but lower quality.",
+  },
 };

frontend/src/hooks/usePipelines.ts

@@ -23,6 +23,19 @@ export function usePipelines() {
         // Transform to camelCase for TypeScript conventions
         const transformed: Record<string, PipelineInfo> = {};
         for (const [id, schema] of Object.entries(schemas.pipelines)) {
+          // Extract VAE types from JSON schema if vae_type field exists
+          // Pydantic v2 represents enum fields using $ref to definitions
+          let vaeTypes: string[] | undefined = undefined;
+          const vaeTypeProperty = schema.config_schema?.properties?.vae_type;
+          if (vaeTypeProperty?.$ref && schema.config_schema?.$defs) {
+            const refPath = vaeTypeProperty.$ref;
+            const defName = refPath.split("/").pop();
+            const definition = schema.config_schema.$defs[defName || ""];
+            if (definition && Array.isArray(definition.enum)) {
+              vaeTypes = definition.enum as string[];
+            }
+          }
+
           transformed[id] = {
             name: schema.name,
             about: schema.description,
@@ -45,6 +58,7 @@ export function usePipelines() {
             recommendedQuantizationVramThreshold:
               schema.recommended_quantization_vram_threshold ?? undefined,
             modified: schema.modified,
+            vaeTypes,
           };
         }
 

frontend/src/lib/api.ts

@@ -356,13 +356,16 @@ export interface PipelineSchemaProperty {
   maximum?: number;
   items?: unknown;
   anyOf?: unknown[];
+  enum?: unknown[];
+  $ref?: string;
 }
 
 export interface PipelineConfigSchema {
   type: string;
   properties: Record<string, PipelineSchemaProperty>;
   required?: string[];
   title?: string;
+  $defs?: Record<string, { enum?: unknown[] }>;
 }
 
 // Mode-specific default overrides

frontend/src/pages/StreamPage.tsx

@@ -21,6 +21,7 @@ import type {
   LoRAConfig,
   LoraMergeStrategy,
   DownloadProgress,
+  VaeType,
 } from "../types";
 import type { PromptItem, PromptTransition } from "../lib/api";
 import { checkModelStatus, downloadPipelineModels } from "../lib/api";
@@ -453,6 +454,11 @@ export function StreamPage() {
     // Note: This setting requires pipeline reload, so we don't send parameter update here
   };
 
+  const handleVaeTypeChange = (vaeType: VaeType) => {
+    updateSettings({ vaeType });
+    // Note: This setting requires pipeline reload, so we don't send parameter update here
+  };
+
   const handleKvCacheAttentionBiasChange = (bias: number) => {
     updateSettings({ kvCacheAttentionBias: bias });
     // Send KV cache attention bias update to backend
@@ -725,6 +731,7 @@ export function StreamPage() {
         if (currentPipeline?.supportsQuantization) {
           loadParams.seed = settings.seed ?? 42;
           loadParams.quantization = settings.quantization ?? null;
+          loadParams.vae_type = settings.vaeType ?? "wan";
         }
 
         // Add LoRA parameters if pipeline supports LoRA
@@ -1117,6 +1124,9 @@ export function StreamPage() {
             onVaceEnabledChange={handleVaceEnabledChange}
             vaceContextScale={settings.vaceContextScale ?? 1.0}
             onVaceContextScaleChange={handleVaceContextScaleChange}
+            vaeType={settings.vaeType ?? "wan"}
+            onVaeTypeChange={handleVaeTypeChange}
+            vaeTypes={pipelines?.[settings.pipelineId]?.vaeTypes ?? ["wan"]}
           />
         </div>
       </div>

frontend/src/types/index.ts

@@ -4,6 +4,9 @@ export type PipelineId = string;
 // Input mode for pipeline operation
 export type InputMode = "text" | "video";
 
+// VAE type for model selection (dynamic from backend registry)
+export type VaeType = string;
+
 // WebRTC ICE server configuration
 export interface IceServerConfig {
   urls: string | string[];
@@ -73,6 +76,8 @@ export interface SettingsState {
   vaceEnabled?: boolean;
   refImages?: string[];
   vaceContextScale?: number;
+  // VAE type selection
+  vaeType?: VaeType;
 }
 
 export interface PipelineInfo {
@@ -100,6 +105,8 @@ export interface PipelineInfo {
   supportsQuantization?: boolean;
   minDimension?: number;
   recommendedQuantizationVramThreshold?: number | null;
+  // Available VAE types from config schema enum (derived from vae_type field presence)
+  vaeTypes?: string[];
 }
 
 export interface DownloadProgress {

src/scope/core/pipelines/krea_realtime_video/pipeline.py

@@ -18,7 +18,7 @@
 from ..utils import Quantization, load_model_config, validate_resolution
 from ..wan2_1.components import WanDiffusionWrapper, WanTextEncoderWrapper
 from ..wan2_1.lora.mixin import LoRAEnabledPipeline
-from ..wan2_1.vae import WanVAEWrapper
+from ..wan2_1.vae import create_vae
 from .modular_blocks import KreaRealtimeVideoBlocks
 from .schema import KreaRealtimeVideoConfig
 
@@ -128,12 +128,16 @@ def __init__(
         # Move text encoder to target device but use dtype of weights
         text_encoder = text_encoder.to(device=device)
 
-        # Load vae
+        # Load VAE using create_vae factory (supports multiple VAE types)
+        vae_type = getattr(config, "vae_type", "wan")
         start = time.time()
-        vae = WanVAEWrapper(
-            model_name=base_model_name, model_dir=model_dir, vae_path=vae_path
+        vae = create_vae(
+            model_dir=model_dir,
+            model_name=base_model_name,
+            vae_type=vae_type,
+            vae_path=vae_path,
         )
-        print(f"Loaded VAE in {time.time() - start:.3f}s")
+        print(f"Loaded VAE (type={vae_type}) in {time.time() - start:.3f}s")
         # Move VAE to target device and use target dtype
         vae = vae.to(device=device, dtype=dtype)
 

src/scope/core/pipelines/krea_realtime_video/schema.py

@@ -1,4 +1,7 @@
+from pydantic import Field
+
 from ..base_schema import BasePipelineConfig, ModeDefaults
+from ..utils import VaeType
 
 
 class KreaRealtimeVideoConfig(BasePipelineConfig):
@@ -26,6 +29,10 @@ class KreaRealtimeVideoConfig(BasePipelineConfig):
     height: int = 320
     width: int = 576
     denoising_steps: list[int] = [1000, 750, 500, 250]
+    vae_type: VaeType = Field(
+        default=VaeType.WAN,
+        description="VAE type to use. 'wan' is the full VAE, 'lightvae' is 75% pruned (faster but lower quality).",
+    )
 
     modes = {
         "text": ModeDefaults(default=True),

src/scope/core/pipelines/longlive/pipeline.py

@@ -20,7 +20,7 @@
 from ..wan2_1.lora.mixin import LoRAEnabledPipeline
 from ..wan2_1.lora.strategies.module_targeted_lora import ModuleTargetedLoRAStrategy
 from ..wan2_1.vace import VACEEnabledPipeline
-from ..wan2_1.vae import WanVAEWrapper
+from ..wan2_1.vae import create_vae
 from .modular_blocks import LongLiveBlocks
 from .schema import LongLiveConfig
 
@@ -145,10 +145,13 @@ def __init__(
         # Move text encoder to target device but use dtype of weights
         text_encoder = text_encoder.to(device=device)
 
-        # Load VAE using unified WanVAEWrapper
+        # Load VAE using create_vae factory (supports multiple VAE types)
+        vae_type = getattr(config, "vae_type", "wan")
         start = time.time()
-        vae = WanVAEWrapper(model_dir=model_dir, model_name=base_model_name)
-        print(f"Loaded VAE in {time.time() - start:.3f}s")
+        vae = create_vae(
+            model_dir=model_dir, model_name=base_model_name, vae_type=vae_type
+        )
+        print(f"Loaded VAE (type={vae_type}) in {time.time() - start:.3f}s")
         # Move VAE to target device and use target dtype
         vae = vae.to(device=device, dtype=dtype)
 

src/scope/core/pipelines/longlive/schema.py

@@ -1,4 +1,7 @@
+from pydantic import Field
+
 from ..base_schema import BasePipelineConfig, ModeDefaults
+from ..utils import VaeType
 
 
 class LongLiveConfig(BasePipelineConfig):
@@ -23,6 +26,10 @@ class LongLiveConfig(BasePipelineConfig):
     height: int = 320
     width: int = 576
     denoising_steps: list[int] = [1000, 750, 500, 250]
+    vae_type: VaeType = Field(
+        default=VaeType.WAN,
+        description="VAE type to use. 'wan' is the full VAE, 'lightvae' is 75% pruned (faster but lower quality).",
+    )
 
     modes = {
         "text": ModeDefaults(default=True),

src/scope/core/pipelines/streamdiffusionv2/schema.py

@@ -1,4 +1,7 @@
+from pydantic import Field
+
 from ..base_schema import BasePipelineConfig, ModeDefaults
+from ..utils import VaeType
 
 
 class StreamDiffusionV2Config(BasePipelineConfig):
@@ -24,6 +27,10 @@ class StreamDiffusionV2Config(BasePipelineConfig):
     noise_scale: float = 0.7
     noise_controller: bool = True
     input_size: int = 4
+    vae_type: VaeType = Field(
+        default=VaeType.WAN,
+        description="VAE type to use. 'wan' is the full VAE, 'lightvae' is 75% pruned (faster but lower quality).",
+    )
 
     modes = {
         "text": ModeDefaults(

src/scope/core/pipelines/utils.py

@@ -14,6 +14,13 @@ class Quantization(str, Enum):
     FP8_E4M3FN = "fp8_e4m3fn"
 
 
+class VaeType(str, Enum):
+    """VAE type enumeration."""
+
+    WAN = "wan"
+    LIGHTVAE = "lightvae"
+
+
 def load_state_dict(weights_path: str) -> dict:
     """Load weights with automatic format detection."""
     if not os.path.exists(weights_path):