fix: detect first valid SSE chunk and refresh server props once

Author: ServeurpersoCom

tools/server/webui/src/lib/services/chat.ts

@@ -56,6 +56,7 @@ export class ChatService {
 			onReasoningChunk,
 			onToolCallChunk,
 			onModel,
+			onFirstValidChunk,
 			// Generation parameters
 			temperature,
 			max_tokens,
@@ -206,6 +207,7 @@ export class ChatService {
 					onReasoningChunk,
 					onToolCallChunk,
 					onModel,
+					onFirstValidChunk,
 					conversationId,
 					abortController.signal
 				);
@@ -274,6 +276,7 @@ export class ChatService {
 		onReasoningChunk?: (chunk: string) => void,
 		onToolCallChunk?: (chunk: string) => void,
 		onModel?: (model: string) => void,
+		onFirstValidChunk?: () => void,
 		conversationId?: string,
 		abortSignal?: AbortSignal
 	): Promise<void> {
@@ -293,6 +296,7 @@ export class ChatService {
 		let lastTimings: ChatMessageTimings | undefined;
 		let streamFinished = false;
 		let modelEmitted = false;
+		let firstValidChunkEmitted = false;
 
 		const finalizeOpenToolCallBatch = () => {
 			if (!hasOpenToolCallBatch) {
@@ -369,6 +373,14 @@ export class ChatService {
 							onModel?.(chunkModel);
 						}
 
+						if (!firstValidChunkEmitted && parsed.object === 'chat.completion.chunk') {
+							firstValidChunkEmitted = true;
+
+							if (!abortSignal?.aborted) {
+								onFirstValidChunk?.();
+							}
+						}
+
 						const content = parsed.choices[0]?.delta?.content;
 						const reasoningContent = parsed.choices[0]?.delta?.reasoning_content;
 						const toolCalls = parsed.choices[0]?.delta?.tool_calls;

tools/server/webui/src/lib/stores/chat.svelte.ts

@@ -365,77 +365,31 @@ class ChatStore {
 
 		let resolvedModel: string | null = null;
 		let modelPersisted = false;
-		const PROPS_REFRESH_RETRY_DELAY_MS = 1_000;
-		let serverPropsRefreshRequested = false;
-		let lastPropsRefreshAttempt = 0;
 		const currentConfig = config();
 		const preferServerPropsModel = !currentConfig.modelSelectorEnabled;
+		let serverPropsRefreshed = false;
+		let updateModelFromServerProps: ((persistImmediately?: boolean) => void) | null = null;
 
-		const resetPropsRefreshGate = (options?: { immediate?: boolean }) => {
-			serverPropsRefreshRequested = false;
-			if (options?.immediate) {
-				lastPropsRefreshAttempt = Date.now() - PROPS_REFRESH_RETRY_DELAY_MS;
-			} else {
-				lastPropsRefreshAttempt = Date.now();
-			}
-		};
-
-		const ensureServerPropsRefresh = () => {
-			const now = Date.now();
-
-			if (serverPropsRefreshRequested) {
-				if (resolvedModel) {
-					const currentModel = serverStore.modelName;
-					const normalizedStoreModel = currentModel ? normalizeModelName(currentModel) : null;
-
-					if (!normalizedStoreModel || normalizedStoreModel !== resolvedModel) {
-						resetPropsRefreshGate({ immediate: true });
-					} else {
-						return;
-					}
-				} else {
-					return;
-				}
-			}
-
-			if (now - lastPropsRefreshAttempt < PROPS_REFRESH_RETRY_DELAY_MS) {
+		const refreshServerPropsOnce = () => {
+			if (serverPropsRefreshed) {
 				return;
 			}
 
-			serverPropsRefreshRequested = true;
-			lastPropsRefreshAttempt = now;
+			serverPropsRefreshed = true;
 
 			const hasExistingProps = serverStore.serverProps !== null;
 
 			serverStore
 				.fetchServerProps({ silent: hasExistingProps })
 				.then(() => {
-					if (!resolvedModel) {
-						return;
-					}
-
-					const currentModel = serverStore.modelName;
-
-					if (!currentModel) {
-						resetPropsRefreshGate({ immediate: true });
-						return;
-					}
-
-					const normalizedStoreModel = normalizeModelName(currentModel);
-
-					if (!normalizedStoreModel || normalizedStoreModel !== resolvedModel) {
-						resetPropsRefreshGate({ immediate: true });
-					}
+					updateModelFromServerProps?.(true);
 				})
 				.catch((error) => {
-					console.error('Failed to refresh server props during streaming:', error);
-					resetPropsRefreshGate();
+					console.warn('Failed to refresh server props after streaming started:', error);
 				});
 		};
 
 		const recordModel = (modelName: string | null | undefined, persistImmediately = true): void => {
-			ensureServerPropsRefresh();
-
 			const serverModelName = serverStore.modelName;
 			const preferredModelSource = preferServerPropsModel
 				? (serverModelName ?? modelName ?? null)
@@ -470,9 +424,7 @@ class ChatStore {
 		};
 
 		if (preferServerPropsModel) {
-			const hasExistingProps = serverStore.serverProps !== null;
-
-			const updateModelFromServerProps = (persistImmediately = true) => {
+			updateModelFromServerProps = (persistImmediately = true) => {
 				const currentServerModel = serverStore.modelName;
 
 				if (!currentServerModel) {
@@ -483,15 +435,6 @@ class ChatStore {
 			};
 
 			updateModelFromServerProps(false);
-
-			serverStore
-				.fetchServerProps({ silent: hasExistingProps })
-				.then(() => {
-					updateModelFromServerProps(true);
-				})
-				.catch((error) => {
-					console.warn('Failed to fetch server props before streaming:', error);
-				});
 		}
 
 		slotsService.startStreaming();
@@ -502,9 +445,10 @@ class ChatStore {
 			{
 				...this.getApiOptions(),
 
+				onFirstValidChunk: () => {
+					refreshServerPropsOnce();
+				},
 				onChunk: (chunk: string) => {
-					ensureServerPropsRefresh();
-
 					streamedContent += chunk;
 					this.setConversationStreaming(
 						assistantMessage.convId,
@@ -519,8 +463,6 @@ class ChatStore {
 				},
 
 				onReasoningChunk: (reasoningChunk: string) => {
-					ensureServerPropsRefresh();
-
 					streamedReasoningContent += reasoningChunk;
 
 					const messageIndex = this.findMessageIndex(assistantMessage.id);
@@ -533,8 +475,6 @@ class ChatStore {
 				},
 
 				onToolCallChunk: (toolCallChunk: string) => {
-					ensureServerPropsRefresh();
-
 					const chunk = toolCallChunk.trim();
 
 					if (!chunk) {
@@ -554,8 +494,6 @@ class ChatStore {
 					timings?: ChatMessageTimings,
 					toolCallContent?: string
 				) => {
-					ensureServerPropsRefresh();
-
 					slotsService.stopStreaming();
 
 					const updateData: {

tools/server/webui/src/lib/types/api.d.ts

@@ -202,6 +202,7 @@ export interface ApiChatCompletionToolCall extends ApiChatCompletionToolCallDelt
 }
 
 export interface ApiChatCompletionStreamChunk {
+	object?: string;
 	model?: string;
 	choices: Array<{
 		model?: string;

tools/server/webui/src/lib/types/settings.d.ts

@@ -43,6 +43,7 @@ export interface SettingsChatServiceOptions {
 	onReasoningChunk?: (chunk: string) => void;
 	onToolCallChunk?: (chunk: string) => void;
 	onModel?: (model: string) => void;
+	onFirstValidChunk?: () => void;
 	onComplete?: (
 		response: string,
 		reasoningContent?: string,