webui: remove client-side context pre-check and rely on backend for limits (ggml-org#16506)

* fix: make SSE client robust to premature [DONE] in agentic proxy chains

* webui: remove client-side context pre-check and rely on backend for limits

Removed the client-side context window pre-check and now simply sends messages
while keeping the dialog imports limited to core components, eliminating the
maximum context alert path

Simplified streaming and non-streaming chat error handling to surface a generic
'No response received from server' error whenever the backend returns no content

Removed the obsolete maxContextError plumbing from the chat store so state
management now focuses on the core message flow without special context-limit cases

* webui: cosmetic rename of error messages

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

Co-authored-by: Aleksander Grygier <[email protected]>

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

Co-authored-by: Aleksander Grygier <[email protected]>

* Update tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreen.svelte

Co-authored-by: Aleksander Grygier <[email protected]>

* Update tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreen.svelte

Co-authored-by: Aleksander Grygier <[email protected]>

* chore: update webui build output

---------

Co-authored-by: Aleksander Grygier <[email protected]>

Author: ServeurpersoCom

tools/server/public/index.html.gz

@@ -7,6 +7,7 @@
 		ChatMessages,
 		ChatProcessingInfo,
 		EmptyFileAlertDialog,
+		ChatErrorDialog,
 		ServerErrorSplash,
 		ServerInfo,
 		ServerLoadingSplash,
@@ -22,10 +23,11 @@
 		activeMessages,
 		activeConversation,
 		deleteConversation,
+		dismissErrorDialog,
+		errorDialog,
 		isLoading,
 		sendMessage,
-		stopGeneration,
-		setMaxContextError
+		stopGeneration
 	} from '$lib/stores/chat.svelte';
 	import {
 		supportsVision,
@@ -34,7 +36,6 @@
 		serverWarning,
 		serverStore
 	} from '$lib/stores/server.svelte';
-	import { contextService } from '$lib/services';
 	import { parseFilesToMessageExtras } from '$lib/utils/convert-files-to-extra';
 	import { isFileTypeSupported } from '$lib/utils/file-type';
 	import { filterFilesByModalities } from '$lib/utils/modality-file-validation';
@@ -79,6 +80,7 @@
 		showCenteredEmpty && !activeConversation() && activeMessages().length === 0 && !isLoading()
 	);
 
+	let activeErrorDialog = $derived(errorDialog());
 	let isServerLoading = $derived(serverLoading());
 
 	async function handleDeleteConfirm() {
@@ -105,6 +107,12 @@
 		}
 	}
 
+	function handleErrorDialogOpenChange(open: boolean) {
+		if (!open) {
+			dismissErrorDialog();
+		}
+	}
+
 	function handleDragOver(event: DragEvent) {
 		event.preventDefault();
 	}
@@ -183,21 +191,6 @@
 
 		const extras = result?.extras;
 
-		// Check context limit using real-time slots data
-		const contextCheck = await contextService.checkContextLimit();
-
-		if (contextCheck && contextCheck.wouldExceed) {
-			const errorMessage = contextService.getContextErrorMessage(contextCheck);
-
-			setMaxContextError({
-				message: errorMessage,
-				estimatedTokens: contextCheck.currentUsage,
-				maxContext: contextCheck.maxContext
-			});
-
-			return false;
-		}
-
 		// Enable autoscroll for user-initiated message sending
 		userScrolledUp = false;
 		autoScrollEnabled = true;
@@ -461,6 +454,13 @@
 	}}
 />
 
+<ChatErrorDialog
+	message={activeErrorDialog?.message ?? ''}
+	onOpenChange={handleErrorDialogOpenChange}
+	open={Boolean(activeErrorDialog)}
+	type={activeErrorDialog?.type ?? 'server'}
+/>
+
 <style>
 	.conversation-chat-form {
 		position: relative;

tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreen.svelte

@@ -0,0 +1,60 @@
+<script lang="ts">
+	import * as AlertDialog from '$lib/components/ui/alert-dialog';
+	import { AlertTriangle, TimerOff } from '@lucide/svelte';
+
+	interface Props {
+		open: boolean;
+		type: 'timeout' | 'server';
+		message: string;
+		onOpenChange?: (open: boolean) => void;
+	}
+
+	let { open = $bindable(), type, message, onOpenChange }: Props = $props();
+
+	const isTimeout = $derived(type === 'timeout');
+	const title = $derived(isTimeout ? 'TCP Timeout' : 'Server Error');
+	const description = $derived(
+		isTimeout
+			? 'The request did not receive a response from the server before timing out.'
+			: 'The server responded with an error message. Review the details below.'
+	);
+	const iconClass = $derived(isTimeout ? 'text-destructive' : 'text-amber-500');
+	const badgeClass = $derived(
+		isTimeout
+			? 'border-destructive/40 bg-destructive/10 text-destructive'
+			: 'border-amber-500/40 bg-amber-500/10 text-amber-600 dark:text-amber-400'
+	);
+
+	function handleOpenChange(newOpen: boolean) {
+		open = newOpen;
+		onOpenChange?.(newOpen);
+	}
+</script>
+
+<AlertDialog.Root {open} onOpenChange={handleOpenChange}>
+	<AlertDialog.Content>
+		<AlertDialog.Header>
+			<AlertDialog.Title class="flex items-center gap-2">
+				{#if isTimeout}
+					<TimerOff class={`h-5 w-5 ${iconClass}`} />
+				{:else}
+					<AlertTriangle class={`h-5 w-5 ${iconClass}`} />
+				{/if}
+
+				{title}
+			</AlertDialog.Title>
+
+			<AlertDialog.Description>
+				{description}
+			</AlertDialog.Description>
+		</AlertDialog.Header>
+
+		<div class={`rounded-lg border px-4 py-3 text-sm ${badgeClass}`}>
+			<p class="font-medium">{message}</p>
+		</div>
+
+		<AlertDialog.Footer>
+			<AlertDialog.Action onclick={() => handleOpenChange(false)}>Close</AlertDialog.Action>
+		</AlertDialog.Footer>
+	</AlertDialog.Content>
+</AlertDialog.Root>

tools/server/webui/src/lib/components/app/dialogs/ChatErrorDialog.svelte

@@ -30,12 +30,11 @@ export { default as ChatSidebar } from './chat/ChatSidebar/ChatSidebar.svelte';
 export { default as ChatSidebarConversationItem } from './chat/ChatSidebar/ChatSidebarConversationItem.svelte';
 export { default as ChatSidebarSearch } from './chat/ChatSidebar/ChatSidebarSearch.svelte';
 
+export { default as ChatErrorDialog } from './dialogs/ChatErrorDialog.svelte';
 export { default as EmptyFileAlertDialog } from './dialogs/EmptyFileAlertDialog.svelte';
 
 export { default as ConversationTitleUpdateDialog } from './dialogs/ConversationTitleUpdateDialog.svelte';
 
-export { default as MaximumContextAlertDialog } from './dialogs/MaximumContextAlertDialog.svelte';
-
 export { default as KeyboardShortcutInfo } from './misc/KeyboardShortcutInfo.svelte';
 
 export { default as MarkdownContent } from './misc/MarkdownContent.svelte';

tools/server/webui/src/lib/components/app/dialogs/MaximumContextAlertDialog.svelte

@@ -13,7 +13,7 @@ import { slotsService } from './slots';
  *   - Manages streaming and non-streaming response parsing
  *   - Provides request abortion capabilities
  *   - Converts database messages to API format
- *   - Handles error translation and context detection
+ *   - Handles error translation for server responses
  *
  * - **ChatStore**: Stateful orchestration and UI state management
  *   - Uses ChatService for all AI model communication
@@ -26,7 +26,6 @@ import { slotsService } from './slots';
  * - Streaming response handling with real-time callbacks
  * - Reasoning content extraction and processing
  * - File attachment processing (images, PDFs, audio, text)
- * - Context error detection and reporting
  * - Request lifecycle management (abort, cleanup)
  */
 export class ChatService {
@@ -209,10 +208,13 @@ export class ChatService {
 					userFriendlyError = new Error(
 						'Unable to connect to server - please check if the server is running'
 					);
+					userFriendlyError.name = 'NetworkError';
 				} else if (error.message.includes('ECONNREFUSED')) {
 					userFriendlyError = new Error('Connection refused - server may be offline');
+					userFriendlyError.name = 'NetworkError';
 				} else if (error.message.includes('ETIMEDOUT')) {
-					userFriendlyError = new Error('Request timeout - server may be overloaded');
+					userFriendlyError = new Error('Request timed out - the server took too long to respond');
+					userFriendlyError.name = 'TimeoutError';
 				} else {
 					userFriendlyError = error;
 				}
@@ -262,6 +264,7 @@ export class ChatService {
 		let fullReasoningContent = '';
 		let hasReceivedData = false;
 		let lastTimings: ChatMessageTimings | undefined;
+		let streamFinished = false;
 
 		try {
 			let chunk = '';
@@ -277,18 +280,8 @@ export class ChatService {
 					if (line.startsWith('data: ')) {
 						const data = line.slice(6);
 						if (data === '[DONE]') {
-							if (!hasReceivedData && aggregatedContent.length === 0) {
-								const contextError = new Error(
-									'The request exceeds the available context size. Try increasing the context size or enable context shift.'
-								);
-								contextError.name = 'ContextError';
-								onError?.(contextError);
-								return;
-							}
-
-							onComplete?.(aggregatedContent, fullReasoningContent || undefined, lastTimings);
-
-							return;
+							streamFinished = true;
+							continue;
 						}
 
 						try {
@@ -326,13 +319,13 @@ export class ChatService {
 				}
 			}
 
-			if (!hasReceivedData && aggregatedContent.length === 0) {
-				const contextError = new Error(
-					'The request exceeds the available context size. Try increasing the context size or enable context shift.'
-				);
-				contextError.name = 'ContextError';
-				onError?.(contextError);
-				return;
+			if (streamFinished) {
+				if (!hasReceivedData && aggregatedContent.length === 0) {
+					const noResponseError = new Error('No response received from server. Please try again.');
+					throw noResponseError;
+				}
+
+				onComplete?.(aggregatedContent, fullReasoningContent || undefined, lastTimings);
 			}
 		} catch (error) {
 			const err = error instanceof Error ? error : new Error('Stream error');
@@ -368,12 +361,8 @@ export class ChatService {
 			const responseText = await response.text();
 
 			if (!responseText.trim()) {
-				const contextError = new Error(
-					'The request exceeds the available context size. Try increasing the context size or enable context shift.'
-				);
-				contextError.name = 'ContextError';
-				onError?.(contextError);
-				throw contextError;
+				const noResponseError = new Error('No response received from server. Please try again.');
+				throw noResponseError;
 			}
 
 			const data: ApiChatCompletionResponse = JSON.parse(responseText);
@@ -385,22 +374,14 @@ export class ChatService {
 			}
 
 			if (!content.trim()) {
-				const contextError = new Error(
-					'The request exceeds the available context size. Try increasing the context size or enable context shift.'
-				);
-				contextError.name = 'ContextError';
-				onError?.(contextError);
-				throw contextError;
+				const noResponseError = new Error('No response received from server. Please try again.');
+				throw noResponseError;
 			}
 
 			onComplete?.(content, reasoningContent);
 
 			return content;
 		} catch (error) {
-			if (error instanceof Error && error.name === 'ContextError') {
-				throw error;
-			}
-
 			const err = error instanceof Error ? error : new Error('Parse error');
 
 			onError?.(err);
@@ -594,37 +575,19 @@ export class ChatService {
 			const errorText = await response.text();
 			const errorData: ApiErrorResponse = JSON.parse(errorText);
 
-			if (errorData.error?.type === 'exceed_context_size_error') {
-				const contextError = errorData.error as ApiContextSizeError;
-				const error = new Error(contextError.message);
-				error.name = 'ContextError';
-				// Attach structured context information
-				(
-					error as Error & {
-						contextInfo?: { promptTokens: number; maxContext: number; estimatedTokens: number };
-					}
-				).contextInfo = {
-					promptTokens: contextError.n_prompt_tokens,
-					maxContext: contextError.n_ctx,
-					estimatedTokens: contextError.n_prompt_tokens
-				};
-				return error;
-			}
-
-			// Fallback for other error types
 			const message = errorData.error?.message || 'Unknown server error';
-			return new Error(message);
+			const error = new Error(message);
+			error.name = response.status === 400 ? 'ServerError' : 'HttpError';
+
+			return error;
 		} catch {
 			// If we can't parse the error response, return a generic error
-			return new Error(`Server error (${response.status}): ${response.statusText}`);
+			const fallback = new Error(`Server error (${response.status}): ${response.statusText}`);
+			fallback.name = 'HttpError';
+			return fallback;
 		}
 	}
 
-	/**
-	 * Updates the processing state with timing information from the server response
-	 * @param timings - Timing data from the API response
-	 * @param promptProgress - Progress data from the API response
-	 */
 	private updateProcessingState(
 		timings?: ChatMessageTimings,
 		promptProgress?: ChatMessagePromptProgress