Merge branch 'upstream' into concedo_experimental

# Conflicts:
#	.devops/intel.Dockerfile
#	.github/workflows/python-type-check.yml
#	embd_res/templates/Qwen3.5-4B.jinja
#	examples/model-conversion/scripts/causal/compare-logits.py
#	examples/model-conversion/scripts/utils/check-nmse.py
#	examples/model-conversion/scripts/utils/compare_tokens.py
#	examples/model-conversion/scripts/utils/semantic_check.py
#	examples/sycl/build.sh
#	examples/sycl/run-llama2.sh
#	ggml/src/ggml-hexagon/htp/flash-attn-ops.c
#	ggml/src/ggml-hexagon/htp/hex-dma.h
#	ggml/src/ggml-hexagon/htp/rope-ops.c
#	scripts/gen-unicode-data.py
#	tests/test-chat.cpp

tools/server/webui/src/lib/components/app/models/ModelId.svelte

@@ -18,7 +18,8 @@
 		showRaw = undefined,
 		aliases,
 		tags,
-		class: className = ''
+		class: className = '',
+		...rest
 	}: Props = $props();
 
 	const badgeClass =
@@ -36,9 +37,9 @@
 </script>
 
 {#if resolvedShowRaw}
-	<TruncatedText class="font-medium {className}" showTooltip={false} text={modelId} />
+	<TruncatedText class="font-medium {className}" showTooltip={false} text={modelId} {...rest} />
 {:else}
-	<span class="flex min-w-0 flex-wrap items-center gap-1 {className}">
+	<span class="flex min-w-0 flex-wrap items-center gap-1 {className}" {...rest}>
 		<span class="min-w-0 truncate font-medium">
 			{#if showOrgName && parsed.orgName && !(aliases && aliases.length > 0)}{parsed.orgName}/{/if}{displayName}
 		</span>

tools/server/webui/src/lib/components/app/models/ModelsSelector.svelte

@@ -271,50 +271,49 @@
 		{#if isRouter}
 			<DropdownMenu.Root bind:open={isOpen} onOpenChange={handleOpenChange}>
 				<DropdownMenu.Trigger
-					disabled={disabled || updating}
-					onclick={(e) => {
-						e.preventDefault();
-						e.stopPropagation();
-					}}
-				>
-					<button
-						type="button"
-						class={cn(
-							`inline-grid cursor-pointer grid-cols-[1fr_auto_1fr] items-center gap-1.5 rounded-sm bg-muted-foreground/10 px-1.5 py-1 text-xs transition hover:text-foreground focus:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:cursor-not-allowed disabled:opacity-60`,
-							!isCurrentModelInCache
-								? 'bg-red-400/10 !text-red-400 hover:bg-red-400/20 hover:text-red-400'
-								: forceForegroundText
+					class={cn(
+						`inline-grid cursor-pointer grid-cols-[1fr_auto_1fr] items-center gap-1.5 rounded-sm bg-muted-foreground/10 px-1.5 py-1 text-xs transition hover:text-foreground focus:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:cursor-not-allowed disabled:opacity-60`,
+						!isCurrentModelInCache
+							? 'bg-red-400/10 !text-red-400 hover:bg-red-400/20 hover:text-red-400'
+							: forceForegroundText
+								? 'text-foreground'
+								: isHighlightedCurrentModelActive
 									? 'text-foreground'
-									: isHighlightedCurrentModelActive
-										? 'text-foreground'
-										: 'text-muted-foreground',
-							isOpen ? 'text-foreground' : ''
-						)}
-						style="max-width: min(calc(100cqw - 9rem), 20rem)"
-						disabled={disabled || updating}
-					>
-						<Package class="h-3.5 w-3.5" />
+									: 'text-muted-foreground',
+						isOpen ? 'text-foreground' : ''
+					)}
+					style="max-width: min(calc(100cqw - 9rem), 20rem)"
+					disabled={disabled || updating}
+				>
+					<Package class="h-3.5 w-3.5" />
 
-						{#if selectedOption}
-							<Tooltip.Root>
-								<Tooltip.Trigger class="min-w-0 overflow-hidden">
-									<ModelId modelId={selectedOption.model} class="min-w-0" showOrgName />
-								</Tooltip.Trigger>
+					{#if selectedOption}
+						<Tooltip.Root>
+							<Tooltip.Trigger>
+								<!-- prevent another nested button element -->
+								{#snippet child({ props })}
+									<ModelId
+										modelId={selectedOption.model}
+										class="min-w-0 overflow-hidden"
+										showOrgName
+										{...props}
+									/>
+								{/snippet}
+							</Tooltip.Trigger>
 
-								<Tooltip.Content>
-									<p class="font-mono">{selectedOption.model}</p>
-								</Tooltip.Content>
-							</Tooltip.Root>
-						{:else}
-							<span class="min-w-0 font-medium">Select model</span>
-						{/if}
+							<Tooltip.Content>
+								<p class="font-mono">{selectedOption.model}</p>
+							</Tooltip.Content>
+						</Tooltip.Root>
+					{:else}
+						<span class="min-w-0 font-medium">Select model</span>
+					{/if}
 
-						{#if updating || isLoadingModel}
-							<Loader2 class="h-3 w-3.5 animate-spin" />
-						{:else}
-							<ChevronDown class="h-3 w-3.5" />
-						{/if}
-					</button>
+					{#if updating || isLoadingModel}
+						<Loader2 class="h-3 w-3.5 animate-spin" />
+					{:else}
+						<ChevronDown class="h-3 w-3.5" />
+					{/if}
 				</DropdownMenu.Trigger>
 
 				<DropdownMenu.Content
@@ -407,8 +406,16 @@
 
 				{#if selectedOption}
 					<Tooltip.Root>
-						<Tooltip.Trigger class="min-w-0 overflow-hidden">
-							<ModelId modelId={selectedOption.model} class="min-w-0" showOrgName />
+						<Tooltip.Trigger>
+							<!-- prevent another nested button element -->
+							{#snippet child({ props })}
+								<ModelId
+									modelId={selectedOption.model}
+									class="min-w-0 overflow-hidden"
+									showOrgName
+									{...props}
+								/>
+							{/snippet}
 						</Tooltip.Trigger>
 
 						<Tooltip.Content>

tools/server/webui/src/lib/services/mcp.service.ts

@@ -39,7 +39,13 @@ import type {
 	MCPResourceContent,
 	MCPReadResourceResult
 } from '$lib/types';
-import { buildProxiedUrl, throwIfAborted, isAbortError, createBase64DataUrl } from '$lib/utils';
+import {
+	buildProxiedUrl,
+	buildProxiedHeaders,
+	throwIfAborted,
+	isAbortError,
+	createBase64DataUrl
+} from '$lib/utils';
 
 interface ToolResultContentItem {
 	type: string;
@@ -118,7 +124,7 @@ export class MCPService {
 		const requestInit: RequestInit = {};
 
 		if (config.headers) {
-			requestInit.headers = config.headers;
+			requestInit.headers = buildProxiedHeaders(config.headers);
 		}
 
 		if (config.credentials) {

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

@@ -28,6 +28,7 @@ import {
 	filterByLeafNodeId,
 	findDescendantMessages,
 	findLeafNode,
+	findMessageById,
 	isAbortError
 } from '$lib/utils';
 import {
@@ -416,7 +417,7 @@ class ChatStore {
 		if (!activeConv) return false;
 		try {
 			const allMessages = await conversationsStore.getConversationMessages(activeConv.id);
-			const systemMessage = allMessages.find((m) => m.id === messageId);
+			const systemMessage = findMessageById(allMessages, messageId);
 			if (!systemMessage || systemMessage.role !== MessageRole.SYSTEM) return false;
 			const rootMessage = allMessages.find((m) => m.type === 'root' && m.parent === null);
 			if (!rootMessage) return false;
@@ -878,7 +879,7 @@ class ChatStore {
 			const msg = conversationsStore.activeMessages[idx];
 			if (msg.role !== MessageRole.ASSISTANT) return;
 			const allMessages = await conversationsStore.getConversationMessages(activeConv.id);
-			const parentMessage = allMessages.find((m) => m.id === msg.parent);
+			const parentMessage = findMessageById(allMessages, msg.parent);
 			if (!parentMessage) return;
 			this.setChatLoading(activeConv.id, true);
 			this.clearChatStreaming(activeConv.id);
@@ -928,7 +929,7 @@ class ChatStore {
 		if (!activeConv)
 			return { totalCount: 0, userMessages: 0, assistantMessages: 0, messageTypes: [] };
 		const allMessages = await conversationsStore.getConversationMessages(activeConv.id);
-		const messageToDelete = allMessages.find((m) => m.id === messageId);
+		const messageToDelete = findMessageById(allMessages, messageId);
 
 		// For system messages, don't count descendants as they will be preserved (reparented to root)
 		if (messageToDelete?.role === MessageRole.SYSTEM) {
@@ -975,7 +976,7 @@ class ChatStore {
 		if (!activeConv) return;
 		try {
 			const allMessages = await conversationsStore.getConversationMessages(activeConv.id);
-			const messageToDelete = allMessages.find((m) => m.id === messageId);
+			const messageToDelete = findMessageById(allMessages, messageId);
 
 			if (!messageToDelete) return;
 
@@ -1024,7 +1025,7 @@ class ChatStore {
 			this.clearChatStreaming(activeConv.id);
 
 			const allMessages = await conversationsStore.getConversationMessages(activeConv.id);
-			const dbMessage = allMessages.find((m) => m.id === messageId);
+			const dbMessage = findMessageById(allMessages, messageId);
 
 			if (!dbMessage) {
 				this.setChatLoading(activeConv.id, false);
@@ -1280,7 +1281,10 @@ class ChatStore {
 
 			let messageIdForResponse: string;
 
-			if (msg.children.length === 0) {
+			const dbMsg = findMessageById(allMessages, msg.id);
+			const hasChildren = dbMsg ? dbMsg.children.length > 0 : msg.children.length > 0;
+
+			if (!hasChildren) {
 				// No responses after this message — update in place instead of branching
 				const updates: Partial<DatabaseMessage> = {
 					content: newContent,

tools/server/webui/src/lib/utils/branching.ts

@@ -17,6 +17,17 @@
 
 import { MessageRole } from '$lib/enums';
 
+/**
+ * Finds a message by its ID in the given messages array.
+ */
+export function findMessageById(
+	messages: readonly DatabaseMessage[],
+	id: string | null | undefined
+): DatabaseMessage | undefined {
+	if (!id) return undefined;
+	return messages.find((m) => m.id === id);
+}
+
 /**
  * Filters messages to get the conversation path from root to a specific leaf node.
  * If the leafNodeId doesn't exist, returns the path with the latest timestamp.

tools/server/webui/src/lib/utils/cors-proxy.ts

@@ -19,6 +19,21 @@ export function buildProxiedUrl(targetUrl: string): URL {
 	return proxyUrl;
 }
 
+/**
+ * Wrap original headers for proxying through the CORS proxy. This avoids issues with duplicated llama.cpp-specific and target headers when using the CORS proxy.
+ * @param headers - The original headers to be proxied to target
+ * @returns List of "wrapped" headers to be sent to the CORS proxy
+ */
+export function buildProxiedHeaders(headers: Record<string, string>): Record<string, string> {
+	const proxiedHeaders: Record<string, string> = {};
+
+	for (const [key, value] of Object.entries(headers)) {
+		proxiedHeaders[`X-Proxy-Header-${key}`] = value;
+	}
+
+	return proxiedHeaders;
+}
+
 /**
  * Get a proxied URL string for use in fetch requests.
  * @param targetUrl - The original URL to proxy

tools/server/webui/src/lib/utils/index.ts

@@ -22,6 +22,7 @@ export { default as autoResizeTextarea } from './autoresize-textarea';
 // Branching utilities
 export {
 	filterByLeafNodeId,
+	findMessageById,
 	findLeafNode,
 	findDescendantMessages,
 	getMessageSiblings,
@@ -38,7 +39,7 @@ export { highlightCode, detectIncompleteCodeBlock, type IncompleteCodeBlock } fr
 export { setConfigValue, getConfigValue, configToParameterRecord } from './config-helpers';
 
 // CORS Proxy
-export { buildProxiedUrl, getProxiedUrlString } from './cors-proxy';
+export { buildProxiedUrl, getProxiedUrlString, buildProxiedHeaders } from './cors-proxy';
 
 // Conversation utilities
 export { createMessageCountMap, getMessageCount } from './conversation-utils';