From ac17d9cb28d193bd3a03b66c9740ec2768b48350 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Thu, 2 Oct 2025 15:00:40 +0200
Subject: [PATCH 01/10] refactor: unify reasoning handling via backend
 reasoning_content, drop frontend tag parsing

- Updated the chat message component to surface backend-supplied reasoning via message.thinking while showing the raw assistant content without inline tag scrubbing
- Simplified chat streaming to append content chunks directly, stream reasoning into the message model, and persist any partial reasoning when generation stops
- Refactored the chat service SSE handler to rely on server-provided reasoning_content, removing legacy <think> parsing logic
- Refreshed Storybook data and streaming flows to populate the thinking field explicitly for static and streaming assistant messages
---
 .../app/chat/ChatMessages/ChatMessage.svelte  |  19 +--
 tools/server/webui/src/lib/services/chat.ts   |  96 ++----------
 .../webui/src/lib/stores/chat.svelte.ts       |  19 ++-
 tools/server/webui/src/lib/utils/thinking.ts  | 143 ------------------
 .../src/stories/ChatMessage.stories.svelte    |  50 +++---
 5 files changed, 51 insertions(+), 276 deletions(-)
 delete mode 100644 tools/server/webui/src/lib/utils/thinking.ts

diff --git a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
index c923bf9e040de..83bde07abf6da 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
@@ -1,7 +1,6 @@
 <script lang="ts">
 	import { getDeletionInfo } from '$lib/stores/chat.svelte';
 	import { copyToClipboard } from '$lib/utils/copy';
-	import { parseThinkingContent } from '$lib/utils/thinking';
 	import ChatMessageAssistant from './ChatMessageAssistant.svelte';
 	import ChatMessageUser from './ChatMessageUser.svelte';
 
@@ -47,25 +46,13 @@
 
 	let thinkingContent = $derived.by(() => {
 		if (message.role === 'assistant') {
-			if (message.thinking) {
-				return message.thinking;
-			}
-
-			const parsed = parseThinkingContent(message.content);
-
-			return parsed.thinking;
+			const trimmedThinking = message.thinking?.trim();
+			return trimmedThinking ? trimmedThinking : null;
 		}
 		return null;
 	});
 
-	let messageContent = $derived.by(() => {
-		if (message.role === 'assistant') {
-			const parsed = parseThinkingContent(message.content);
-			return parsed.cleanContent?.replace('<|channel|>analysis', '');
-		}
-
-		return message.content?.replace('<|channel|>analysis', '');
-	});
+	let messageContent = $derived.by(() => message.content);
 
 	function handleCancelEdit() {
 		isEditing = false;
diff --git a/tools/server/webui/src/lib/services/chat.ts b/tools/server/webui/src/lib/services/chat.ts
index 369cdf4e8b935..90df8fa730ba1 100644
--- a/tools/server/webui/src/lib/services/chat.ts
+++ b/tools/server/webui/src/lib/services/chat.ts
@@ -256,10 +256,8 @@ export class ChatService {
 		}
 
 		const decoder = new TextDecoder();
-		let fullResponse = '';
+		let aggregatedContent = '';
 		let fullReasoningContent = '';
-		let regularContent = '';
-		let insideThinkTag = false;
 		let hasReceivedData = false;
 		let lastTimings: ChatMessageTimings | undefined;
 
@@ -277,7 +275,7 @@ export class ChatService {
 					if (line.startsWith('data: ')) {
 						const data = line.slice(6);
 						if (data === '[DONE]') {
-							if (!hasReceivedData && fullResponse.length === 0) {
+							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.'
 								);
@@ -286,7 +284,7 @@ export class ChatService {
 								return;
 							}
 
-							onComplete?.(regularContent, fullReasoningContent || undefined, lastTimings);
+							onComplete?.(aggregatedContent, fullReasoningContent || undefined, lastTimings);
 
 							return;
 						}
@@ -308,36 +306,17 @@ export class ChatService {
 								}
 							}
 
-							if (content) {
-								hasReceivedData = true;
-								fullResponse += content;
-
-								// Track the regular content before processing this chunk
-								const regularContentBefore = regularContent;
-
-								// Process content character by character to handle think tags
-								insideThinkTag = this.processContentForThinkTags(
-									content,
-									insideThinkTag,
-									() => {
-										// Think content is ignored - we don't include it in API requests
-									},
-									(regularChunk) => {
-										regularContent += regularChunk;
-									}
-								);
-
-								const newRegularContent = regularContent.slice(regularContentBefore.length);
-								if (newRegularContent) {
-									onChunk?.(newRegularContent);
-								}
-							}
-
-							if (reasoningContent) {
-								hasReceivedData = true;
-								fullReasoningContent += reasoningContent;
-								onReasoningChunk?.(reasoningContent);
-							}
+                                                        if (content) {
+                                                                hasReceivedData = true;
+                                                                aggregatedContent += content;
+                                                                onChunk?.(content);
+                                                        }
+
+                                                        if (reasoningContent) {
+                                                                hasReceivedData = true;
+                                                                fullReasoningContent += reasoningContent;
+                                                                onReasoningChunk?.(reasoningContent);
+                                                        }
 						} catch (e) {
 							console.error('Error parsing JSON chunk:', e);
 						}
@@ -345,7 +324,7 @@ export class ChatService {
 				}
 			}
 
-			if (!hasReceivedData && fullResponse.length === 0) {
+			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.'
 				);
@@ -552,51 +531,6 @@ export class ChatService {
 		}
 	}
 
-	/**
-	 * Processes content to separate thinking tags from regular content.
-	 * Parses <think> and </think> tags to route content to appropriate handlers.
-	 *
-	 * @param content - The content string to process
-	 * @param currentInsideThinkTag - Current state of whether we're inside a think tag
-	 * @param addThinkContent - Callback to handle content inside think tags
-	 * @param addRegularContent - Callback to handle regular content outside think tags
-	 * @returns Boolean indicating if we're still inside a think tag after processing
-	 * @private
-	 */
-	private processContentForThinkTags(
-		content: string,
-		currentInsideThinkTag: boolean,
-		addThinkContent: (chunk: string) => void,
-		addRegularContent: (chunk: string) => void
-	): boolean {
-		let i = 0;
-		let insideThinkTag = currentInsideThinkTag;
-
-		while (i < content.length) {
-			if (!insideThinkTag && content.substring(i, i + 7) === '<think>') {
-				insideThinkTag = true;
-				i += 7; // Skip the <think> tag
-				continue;
-			}
-
-			if (insideThinkTag && content.substring(i, i + 8) === '</think>') {
-				insideThinkTag = false;
-				i += 8; // Skip the </think> tag
-				continue;
-			}
-
-			if (insideThinkTag) {
-				addThinkContent(content[i]);
-			} else {
-				addRegularContent(content[i]);
-			}
-
-			i++;
-		}
-
-		return insideThinkTag;
-	}
-
 	/**
 	 * Aborts any ongoing chat completion request.
 	 * Cancels the current request and cleans up the abort controller.
diff --git a/tools/server/webui/src/lib/stores/chat.svelte.ts b/tools/server/webui/src/lib/stores/chat.svelte.ts
index 3f9226efeb3b4..e839d8693c683 100644
--- a/tools/server/webui/src/lib/stores/chat.svelte.ts
+++ b/tools/server/webui/src/lib/stores/chat.svelte.ts
@@ -5,7 +5,6 @@ import { config } from '$lib/stores/settings.svelte';
 import { filterByLeafNodeId, findLeafNode, findDescendantMessages } from '$lib/utils/branching';
 import { browser } from '$app/environment';
 import { goto } from '$app/navigation';
-import { extractPartialThinking } from '$lib/utils/thinking';
 
 /**
  * ChatStore - Central state management for chat conversations and AI interactions
@@ -341,12 +340,9 @@ class ChatStore {
 				streamedContent += chunk;
 				this.currentResponse = streamedContent;
 
-				captureModelIfNeeded();
-
-				const partialThinking = extractPartialThinking(streamedContent);
 				const messageIndex = this.findMessageIndex(assistantMessage.id);
 				this.updateMessageAtIndex(messageIndex, {
-					content: partialThinking.remainingContent || streamedContent
+					content: streamedContent
 				});
 			},
 
@@ -694,18 +690,18 @@ class ChatStore {
 
 		if (lastMessage && lastMessage.role === 'assistant') {
 			try {
-				const partialThinking = extractPartialThinking(this.currentResponse);
+				const contentToSave = this.currentResponse;
 
 				const updateData: {
 					content: string;
 					thinking?: string;
 					timings?: ChatMessageTimings;
 				} = {
-					content: partialThinking.remainingContent || this.currentResponse
+					content: contentToSave
 				};
 
-				if (partialThinking.thinking) {
-					updateData.thinking = partialThinking.thinking;
+				if (lastMessage.thinking?.trim()) {
+					updateData.thinking = lastMessage.thinking;
 				}
 
 				const lastKnownState = await slotsService.getCurrentState();
@@ -725,7 +721,10 @@ class ChatStore {
 
 				await DatabaseStore.updateMessage(lastMessage.id, updateData);
 
-				lastMessage.content = partialThinking.remainingContent || this.currentResponse;
+				lastMessage.content = contentToSave;
+				if (updateData.thinking !== undefined) {
+					lastMessage.thinking = updateData.thinking;
+				}
 				if (updateData.timings) {
 					lastMessage.timings = updateData.timings;
 				}
diff --git a/tools/server/webui/src/lib/utils/thinking.ts b/tools/server/webui/src/lib/utils/thinking.ts
deleted file mode 100644
index bed13fcecf159..0000000000000
--- a/tools/server/webui/src/lib/utils/thinking.ts
+++ /dev/null
@@ -1,143 +0,0 @@
-/**
- * Parses thinking content from a message that may contain <think> tags or [THINK] tags
- * Returns an object with thinking content and cleaned message content
- * Handles both complete blocks and incomplete blocks (streaming)
- * Supports formats: <think>...</think> and [THINK]...[/THINK]
- * @param content - The message content to parse
- * @returns An object containing the extracted thinking content and the cleaned message content
- */
-export function parseThinkingContent(content: string): {
-	thinking: string | null;
-	cleanContent: string;
-} {
-	const incompleteThinkMatch = content.includes('<think>') && !content.includes('</think>');
-	const incompleteThinkBracketMatch = content.includes('[THINK]') && !content.includes('[/THINK]');
-
-	if (incompleteThinkMatch) {
-		const cleanContent = content.split('</think>')?.[1]?.trim();
-		const thinkingContent = content.split('<think>')?.[1]?.trim();
-
-		return {
-			cleanContent,
-			thinking: thinkingContent
-		};
-	}
-
-	if (incompleteThinkBracketMatch) {
-		const cleanContent = content.split('[/THINK]')?.[1]?.trim();
-		const thinkingContent = content.split('[THINK]')?.[1]?.trim();
-
-		return {
-			cleanContent,
-			thinking: thinkingContent
-		};
-	}
-
-	const completeThinkMatch = content.match(/<think>([\s\S]*?)<\/think>/);
-	const completeThinkBracketMatch = content.match(/\[THINK\]([\s\S]*?)\[\/THINK\]/);
-
-	if (completeThinkMatch) {
-		const thinkingContent = completeThinkMatch[1]?.trim() ?? '';
-		const cleanContent = `${content.slice(0, completeThinkMatch.index ?? 0)}${content.slice(
-			(completeThinkMatch.index ?? 0) + completeThinkMatch[0].length
-		)}`.trim();
-
-		return {
-			thinking: thinkingContent,
-			cleanContent
-		};
-	}
-
-	if (completeThinkBracketMatch) {
-		const thinkingContent = completeThinkBracketMatch[1]?.trim() ?? '';
-		const cleanContent = `${content.slice(0, completeThinkBracketMatch.index ?? 0)}${content.slice(
-			(completeThinkBracketMatch.index ?? 0) + completeThinkBracketMatch[0].length
-		)}`.trim();
-
-		return {
-			thinking: thinkingContent,
-			cleanContent
-		};
-	}
-
-	return {
-		thinking: null,
-		cleanContent: content
-	};
-}
-
-/**
- * Checks if content contains an opening thinking tag (for streaming)
- * Supports both <think> and [THINK] formats
- * @param content - The message content to check
- * @returns True if the content contains an opening thinking tag
- */
-export function hasThinkingStart(content: string): boolean {
-	return (
-		content.includes('<think>') ||
-		content.includes('[THINK]') ||
-		content.includes('<|channel|>analysis')
-	);
-}
-
-/**
- * Checks if content contains a closing thinking tag (for streaming)
- * Supports both </think> and [/THINK] formats
- * @param content - The message content to check
- * @returns True if the content contains a closing thinking tag
- */
-export function hasThinkingEnd(content: string): boolean {
-	return content.includes('</think>') || content.includes('[/THINK]');
-}
-
-/**
- * Extracts partial thinking content during streaming
- * Supports both <think> and [THINK] formats
- * Used when we have opening tag but not yet closing tag
- * @param content - The message content to extract partial thinking from
- * @returns An object containing the extracted partial thinking content and the remaining content
- */
-export function extractPartialThinking(content: string): {
-	thinking: string | null;
-	remainingContent: string;
-} {
-	const thinkStartIndex = content.indexOf('<think>');
-	const thinkEndIndex = content.indexOf('</think>');
-
-	const bracketStartIndex = content.indexOf('[THINK]');
-	const bracketEndIndex = content.indexOf('[/THINK]');
-
-	const useThinkFormat =
-		thinkStartIndex !== -1 && (bracketStartIndex === -1 || thinkStartIndex < bracketStartIndex);
-	const useBracketFormat =
-		bracketStartIndex !== -1 && (thinkStartIndex === -1 || bracketStartIndex < thinkStartIndex);
-
-	if (useThinkFormat) {
-		if (thinkEndIndex === -1) {
-			const thinkingStart = thinkStartIndex + '<think>'.length;
-
-			return {
-				thinking: content.substring(thinkingStart),
-				remainingContent: content.substring(0, thinkStartIndex)
-			};
-		}
-	} else if (useBracketFormat) {
-		if (bracketEndIndex === -1) {
-			const thinkingStart = bracketStartIndex + '[THINK]'.length;
-
-			return {
-				thinking: content.substring(thinkingStart),
-				remainingContent: content.substring(0, bracketStartIndex)
-			};
-		}
-	} else {
-		return { thinking: null, remainingContent: content };
-	}
-
-	const parsed = parseThinkingContent(content);
-
-	return {
-		thinking: parsed.thinking,
-		remainingContent: parsed.cleanContent
-	};
-}
diff --git a/tools/server/webui/src/stories/ChatMessage.stories.svelte b/tools/server/webui/src/stories/ChatMessage.stories.svelte
index c6377e23cb6fd..e35b3618586d0 100644
--- a/tools/server/webui/src/stories/ChatMessage.stories.svelte
+++ b/tools/server/webui/src/stories/ChatMessage.stories.svelte
@@ -60,7 +60,7 @@
 		children: []
 	});
 
-	// Message with <think> format thinking content
+	// Message with dedicated thinking content
 	const thinkTagMessage: DatabaseMessage = {
 		id: '6',
 		convId: 'conv-1',
@@ -68,13 +68,14 @@
 		timestamp: Date.now() - 1000 * 60 * 2,
 		role: 'assistant',
 		content:
-			"<think>\nLet me analyze this step by step:\n\n1. The user is asking about thinking formats\n2. I need to demonstrate the &lt;think&gt; tag format\n3. This content should be displayed in the thinking section\n4. The main response should be separate\n\nThis is a good example of reasoning content.\n</think>\n\nHere's my response after thinking through the problem. The thinking content above should be displayed separately from this main response content.",
+			"Here's my response after thinking through the problem. The reasoning should appear separately from this main response content.",
 		parent: '1',
-		thinking: '',
+		thinking:
+			"Let me analyze this step by step:\n\n1. The user is asking about thinking formats\n2. I need to demonstrate the &lt;think&gt; tag format\n3. This content should be displayed in the thinking section\n4. The main response should be separate\n\nThis is a good example of reasoning content.",
 		children: []
 	};
 
-	// Message with [THINK] format thinking content
+	// Message with separate DeepSeek-style thinking content
 	const thinkBracketMessage: DatabaseMessage = {
 		id: '7',
 		convId: 'conv-1',
@@ -82,13 +83,14 @@
 		timestamp: Date.now() - 1000 * 60 * 1,
 		role: 'assistant',
 		content:
-			'[THINK]\nThis is the DeepSeek-style thinking format:\n\n- Using square brackets instead of angle brackets\n- Should work identically to the &lt;think&gt; format\n- Content parsing should extract this reasoning\n- Display should be the same as &lt;think&gt; format\n\nBoth formats should be supported seamlessly.\n[/THINK]\n\nThis is the main response content that comes after the [THINK] block. The reasoning above should be parsed and displayed in the thinking section.',
+			"Here's my response after using the [THINK] format to organize the answer. The reasoning content should be shown separately from this response.",
 		parent: '1',
-		thinking: '',
+		thinking:
+			"This is the DeepSeek-style thinking format:\n\n- Using square brackets instead of angle brackets\n- Should work identically to the &lt;think&gt; format\n- Reasoning content should display in the thinking section\n- Main response should be separate\n\nBoth formats should be supported seamlessly.",
 		children: []
 	};
 
-	// Streaming message for <think> format
+	// Streaming message for dedicated thinking content
 	let streamingThinkMessage = $state({
 		id: '8',
 		convId: 'conv-1',
@@ -101,7 +103,7 @@
 		children: []
 	});
 
-	// Streaming message for [THINK] format
+		// Streaming message for DeepSeek-style thinking content
 	let streamingBracketMessage = $state({
 		id: '9',
 		convId: 'conv-1',
@@ -227,28 +229,26 @@
 	}}
 	asChild
 	play={async () => {
-		// Phase 1: Stream <think> reasoning content
+		// Phase 1: Stream reasoning content
 		const thinkingContent =
 			'Let me work through this problem systematically:\n\n1. First, I need to understand what the user is asking\n2. Then I should consider different approaches\n3. I need to evaluate the pros and cons\n4. Finally, I should provide a clear recommendation\n\nThis step-by-step approach will ensure accuracy.';
 
-		let currentContent = '<think>\n';
-		streamingThinkMessage.content = currentContent;
+		streamingThinkMessage.thinking = '';
+		streamingThinkMessage.content = '';
 
 		for (let i = 0; i < thinkingContent.length; i++) {
-			currentContent += thinkingContent[i];
-			streamingThinkMessage.content = currentContent;
+			streamingThinkMessage.thinking += thinkingContent[i];
 			await new Promise((resolve) => setTimeout(resolve, 5));
 		}
 
-		// Close the thinking block
-		currentContent += '\n</think>\n\n';
-		streamingThinkMessage.content = currentContent;
 		await new Promise((resolve) => setTimeout(resolve, 200));
 
 		// Phase 2: Stream main response content
 		const responseContent =
 			"Based on my analysis above, here's the solution:\n\n**Key Points:**\n- The approach should be systematic\n- We need to consider all factors\n- Implementation should be step-by-step\n\nThis ensures the best possible outcome.";
 
+		let currentContent = '';
+
 		for (let i = 0; i < responseContent.length; i++) {
 			currentContent += responseContent[i];
 			streamingThinkMessage.content = currentContent;
@@ -275,27 +275,25 @@
 	}}
 	asChild
 	play={async () => {
-		// Phase 1: Stream [THINK] reasoning content
+		// Phase 1: Stream DeepSeek-style reasoning content
 		const thinkingContent =
-			'Using the DeepSeek format now:\n\n- This demonstrates the &#91;THINK&#93; bracket format\n- Should parse identically to &lt;think&gt; tags\n- The UI should display this in the thinking section\n- Main content should be separate\n\nBoth formats provide the same functionality.';
+			'Using the DeepSeek format now:\n\n- This demonstrates the &#91;THINK&#93; bracket format\n- Should behave identically to the &lt;think&gt; format\n- The UI should display this in the thinking section\n- Main content should be separate\n\nBoth formats provide the same functionality.';
 
-		let currentContent = '[THINK]\n';
-		streamingBracketMessage.content = currentContent;
+		streamingBracketMessage.thinking = '';
+		streamingBracketMessage.content = '';
 
 		for (let i = 0; i < thinkingContent.length; i++) {
-			currentContent += thinkingContent[i];
-			streamingBracketMessage.content = currentContent;
+			streamingBracketMessage.thinking += thinkingContent[i];
 			await new Promise((resolve) => setTimeout(resolve, 5));
 		}
 
-		// Close the thinking block
-		currentContent += '\n[/THINK]\n\n';
-		streamingBracketMessage.content = currentContent;
 		await new Promise((resolve) => setTimeout(resolve, 200));
 
 		// Phase 2: Stream main response content
 		const responseContent =
-			"Here's my response after using the &#91;THINK&#93; format:\n\n**Observations:**\n- Both &lt;think&gt; and &#91;THINK&#93; formats work seamlessly\n- The parsing logic handles both cases\n- UI display is consistent across formats\n\nThis demonstrates the enhanced thinking content support.";
+			"Here's my response after using the &#91;THINK&#93; format:\n\n**Observations:**\n- Both &lt;think&gt; and &#91;THINK&#93; formats work seamlessly\n- The helper logic handles both cases\n- UI display is consistent across formats\n\nThis demonstrates the enhanced thinking content support.";
+
+		let currentContent = '';
 
 		for (let i = 0; i < responseContent.length; i++) {
 			currentContent += responseContent[i];

From 40585f06e390ca1910bb121106fc77964deec96a Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Thu, 2 Oct 2025 22:42:29 +0200
Subject: [PATCH 02/10] refactor: implement streaming-aware universal reasoning
 parser

Remove the streaming mode limitation from --reasoning-format by refactoring
try_parse_reasoning() to handle incremental parsing of <think> tags across
all formats.

- Rework try_parse_reasoning() to track whitespace, partial tags, and
  multiple reasoning segments, allowing proper separation of reasoning_content
  and content in streaming mode
- Parse reasoning tags before tool call handling in content-only and Llama 3.x
  formats to ensure inline <think> blocks are captured correctly
- Change default reasoning_format from 'auto' to 'deepseek' for consistent
  behavior
- Add 'deepseek-legacy' option to preserve old inline behavior when needed
- Update CLI help and documentation to reflect streaming support
- Add parser tests for inline <think>...</think> segments

The parser now continues processing content after </think> closes instead of
stopping, enabling proper message.reasoning_content and message.content
separation in both streaming and non-streaming modes.

Fixes the issue where streaming responses would dump everything (including
post-thinking content) into reasoning_content while leaving content empty.
---
 common/arg.cpp             |   5 +-
 common/chat-parser.cpp     | 115 ++++++++++++++++++++++++++++++++-----
 common/chat.cpp            |   3 +
 common/common.h            |   2 +-
 tests/test-chat-parser.cpp |  28 +++++++++
 tools/server/README.md     |   2 +-
 6 files changed, 138 insertions(+), 17 deletions(-)

diff --git a/common/arg.cpp b/common/arg.cpp
index a020ac4413290..53fc4e39b3654 100644
--- a/common/arg.cpp
+++ b/common/arg.cpp
@@ -3425,8 +3425,9 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
         {"--reasoning-format"}, "FORMAT",
         "controls whether thought tags are allowed and/or extracted from the response, and in which format they're returned; one of:\n"
         "- none: leaves thoughts unparsed in `message.content`\n"
-        "- deepseek: puts thoughts in `message.reasoning_content` (except in streaming mode, which behaves as `none`)\n"
-        "(default: auto)",
+        "- deepseek: puts thoughts in `message.reasoning_content`\n"
+        "- deepseek-legacy: keeps `<think>` tags in `message.content` while also populating `message.reasoning_content`\n"
+        "(default: deepseek)",
         [](common_params & params, const std::string & value) {
             params.reasoning_format = common_reasoning_format_from_name(value);
         }
diff --git a/common/chat-parser.cpp b/common/chat-parser.cpp
index b3362519a68f3..392f0f587b9f1 100644
--- a/common/chat-parser.cpp
+++ b/common/chat-parser.cpp
@@ -3,9 +3,12 @@
 #include "log.h"
 #include "regex-partial.h"
 
+#include <algorithm>
+#include <cctype>
 #include <optional>
 #include <stdexcept>
 #include <string>
+#include <string_view>
 #include <vector>
 
 using json = nlohmann::ordered_json;
@@ -181,25 +184,111 @@ bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think
             add_reasoning_content(stripped_reasoning);
         }
     };
-    if (syntax_.reasoning_format != COMMON_REASONING_FORMAT_NONE) {
-        if (syntax_.thinking_forced_open || try_consume_literal(start_think)) {
-            if (auto res = try_find_literal(end_think)) {
-                handle_reasoning(res->prelude, /* closed */ true);
-                consume_spaces();
-                return true;
-            }
-            auto rest = consume_rest();
+
+    if (syntax_.reasoning_format == COMMON_REASONING_FORMAT_NONE) {
+        return false;
+    }
+
+    const size_t saved_pos = pos_;
+    const size_t saved_content_size = result_.content.size();
+    const size_t saved_reasoning_size = result_.reasoning_content.size();
+
+    auto restore_state = [&]() {
+        move_to(saved_pos);
+        result_.content.resize(saved_content_size);
+        result_.reasoning_content.resize(saved_reasoning_size);
+    };
+
+    // Allow leading whitespace to be preserved as content when reasoning is present at the start
+    size_t cursor = pos_;
+    size_t whitespace_end = cursor;
+    while (whitespace_end < input_.size() && std::isspace(static_cast<unsigned char>(input_[whitespace_end]))) {
+        ++whitespace_end;
+    }
+
+    if (whitespace_end >= input_.size()) {
+        restore_state();
+        if (syntax_.thinking_forced_open) {
+            auto rest = input_.substr(saved_pos);
             if (!rest.empty()) {
                 handle_reasoning(rest, /* closed */ !is_partial());
             }
-            // Allow unclosed thinking tags, for now (https://github.com/ggml-org/llama.cpp/issues/13812, https://github.com/ggml-org/llama.cpp/issues/13877)
-            // if (!syntax_.thinking_forced_open) {
-            //     throw common_chat_msg_partial_exception(end_think);
-            // }
+            move_to(input_.size());
             return true;
         }
+        return false;
+    }
+
+    cursor = whitespace_end;
+    const size_t remaining = input_.size() - cursor;
+    const size_t start_prefix = std::min(start_think.size(), remaining);
+    const bool has_start_tag = input_.compare(cursor, start_prefix, start_think, 0, start_prefix) == 0;
+
+    if (has_start_tag && start_prefix < start_think.size()) {
+        move_to(input_.size());
+        return true;
+    }
+
+    if (has_start_tag) {
+        if (whitespace_end > pos_) {
+            add_content(input_.substr(pos_, whitespace_end - pos_));
+        }
+        cursor += start_think.size();
+    } else if (syntax_.thinking_forced_open) {
+        cursor = whitespace_end;
+    } else {
+        restore_state();
+        return false;
+    }
+    while (true) {
+        if (cursor >= input_.size()) {
+            move_to(input_.size());
+            return true;
+        }
+
+        size_t end_pos = input_.find(end_think, cursor);
+        if (end_pos == std::string::npos) {
+            std::string_view remaining_view(input_.data() + cursor, input_.size() - cursor);
+            size_t partial_off = string_find_partial_stop(remaining_view, end_think);
+            size_t reasoning_end = partial_off == std::string::npos ? input_.size() : cursor + partial_off;
+            if (reasoning_end > cursor) {
+                handle_reasoning(input_.substr(cursor, reasoning_end - cursor), /* closed */ partial_off == std::string::npos && !is_partial());
+            }
+            move_to(input_.size());
+            return true;
+        }
+
+        if (end_pos > cursor) {
+            handle_reasoning(input_.substr(cursor, end_pos - cursor), /* closed */ true);
+        } else {
+            handle_reasoning("", /* closed */ true);
+        }
+
+        cursor = end_pos + end_think.size();
+
+        while (cursor < input_.size() && std::isspace(static_cast<unsigned char>(input_[cursor]))) {
+            ++cursor;
+        }
+
+        const size_t next_remaining = input_.size() - cursor;
+        if (next_remaining == 0) {
+            move_to(cursor);
+            return true;
+        }
+
+        const size_t next_prefix = std::min(start_think.size(), next_remaining);
+        if (input_.compare(cursor, next_prefix, start_think, 0, next_prefix) == 0) {
+            if (next_prefix < start_think.size()) {
+                move_to(input_.size());
+                return true;
+            }
+            cursor += start_think.size();
+            continue;
+        }
+
+        move_to(cursor);
+        return true;
     }
-    return false;
 }
 
 std::string common_chat_msg_parser::consume_rest() {
diff --git a/common/chat.cpp b/common/chat.cpp
index afbb2a2bdd3c4..8587140e1ff0a 100644
--- a/common/chat.cpp
+++ b/common/chat.cpp
@@ -1408,6 +1408,8 @@ static common_chat_params common_chat_params_init_apertus(const common_chat_temp
     return data;
 }
 static void common_chat_parse_llama_3_1(common_chat_msg_parser & builder, bool with_builtin_tools = false) {
+    builder.try_parse_reasoning("<think>", "</think>");
+
     if (!builder.syntax().parse_tool_calls) {
         builder.add_content(builder.consume_rest());
         return;
@@ -2862,6 +2864,7 @@ common_chat_params common_chat_templates_apply(
 }
 
 static void common_chat_parse_content_only(common_chat_msg_parser & builder) {
+    builder.try_parse_reasoning("<think>", "</think>");
     builder.add_content(builder.consume_rest());
 }
 
diff --git a/common/common.h b/common/common.h
index d33788bd100b2..33774dfa44cab 100644
--- a/common/common.h
+++ b/common/common.h
@@ -432,7 +432,7 @@ struct common_params {
     std::string chat_template = "";                                                                         // NOLINT
     bool use_jinja = false;                                                                                 // NOLINT
     bool enable_chat_template = true;
-    common_reasoning_format reasoning_format = COMMON_REASONING_FORMAT_AUTO;
+    common_reasoning_format reasoning_format = COMMON_REASONING_FORMAT_DEEPSEEK;
     int reasoning_budget = -1;
     bool prefill_assistant = true;                                                                          // if true, any trailing assistant message will be prefilled into the response
 
diff --git a/tests/test-chat-parser.cpp b/tests/test-chat-parser.cpp
index 547ebb4871cd4..0b275befb8bf4 100644
--- a/tests/test-chat-parser.cpp
+++ b/tests/test-chat-parser.cpp
@@ -106,6 +106,34 @@ static void test_reasoning() {
     assert_equals("<think>Cogito</think>", builder.result().content);
     assert_equals("Ergo sum", builder.consume_rest());
   }
+  {
+    const std::string variant("content_only_inline_think");
+    common_chat_syntax syntax = {
+        /* .format = */ COMMON_CHAT_FORMAT_CONTENT_ONLY,
+        /* .reasoning_format = */ COMMON_REASONING_FORMAT_DEEPSEEK,
+        /* .reasoning_in_content = */ false,
+        /* .thinking_forced_open = */ false,
+        /* .parse_tool_calls = */ false,
+    };
+    const std::string input = "<think>Pense</think>Bonjour";
+    auto msg = common_chat_parse(input, false, syntax);
+    assert_equals(variant, std::string("Pense"), msg.reasoning_content);
+    assert_equals(variant, std::string("Bonjour"), msg.content);
+  }
+  {
+    const std::string variant("llama_3_inline_think");
+    common_chat_syntax syntax = {
+        /* .format = */ COMMON_CHAT_FORMAT_LLAMA_3_X,
+        /* .reasoning_format = */ COMMON_REASONING_FORMAT_DEEPSEEK,
+        /* .reasoning_in_content = */ false,
+        /* .thinking_forced_open = */ false,
+        /* .parse_tool_calls = */ false,
+    };
+    const std::string input = "<think>Plan</think>Réponse";
+    auto msg = common_chat_parse(input, false, syntax);
+    assert_equals(variant, std::string("Plan"), msg.reasoning_content);
+    assert_equals(variant, std::string("Réponse"), msg.content);
+  }
   // Test DeepSeek V3.1 parsing - reasoning content followed by "</think>" and then regular content
   {
     common_chat_syntax syntax = {
diff --git a/tools/server/README.md b/tools/server/README.md
index 6825c8bf300c6..c47e15c5ade50 100644
--- a/tools/server/README.md
+++ b/tools/server/README.md
@@ -190,7 +190,7 @@ The project is under active development, and we are [looking for feedback and co
 | `--no-slots` | disables slots monitoring endpoint<br/>(env: LLAMA_ARG_NO_ENDPOINT_SLOTS) |
 | `--slot-save-path PATH` | path to save slot kv cache (default: disabled) |
 | `--jinja` | use jinja template for chat (default: disabled)<br/>(env: LLAMA_ARG_JINJA) |
-| `--reasoning-format FORMAT` | controls whether thought tags are allowed and/or extracted from the response, and in which format they're returned; one of:<br/>- none: leaves thoughts unparsed in `message.content`<br/>- deepseek: puts thoughts in `message.reasoning_content` (except in streaming mode, which behaves as `none`)<br/>(default: auto)<br/>(env: LLAMA_ARG_THINK) |
+| `--reasoning-format FORMAT` | controls whether thought tags are allowed and/or extracted from the response, and in which format they're returned; one of:<br/>- none: leaves thoughts unparsed in `message.content`<br/>- deepseek: puts thoughts in `message.reasoning_content`<br/>- deepseek-legacy: keeps `<think>` tags in `message.content` while also populating `message.reasoning_content`<br/>(default: deepseek)<br/>(env: LLAMA_ARG_THINK) |
 | `--reasoning-budget N` | controls the amount of thinking allowed; currently only one of: -1 for unrestricted thinking budget, or 0 to disable thinking (default: -1)<br/>(env: LLAMA_ARG_THINK_BUDGET) |
 | `--chat-template JINJA_TEMPLATE` | set custom jinja chat template (default: template taken from model's metadata)<br/>if suffix/prefix are specified, template will be disabled<br/>only commonly used templates are accepted (unless --jinja is set before this flag):<br/>list of built-in templates:<br/>bailing, chatglm3, chatglm4, chatml, command-r, deepseek, deepseek2, deepseek3, exaone3, exaone4, falcon3, gemma, gigachat, glmedge, gpt-oss, granite, hunyuan-dense, hunyuan-moe, kimi-k2, llama2, llama2-sys, llama2-sys-bos, llama2-sys-strip, llama3, llama4, megrez, minicpm, mistral-v1, mistral-v3, mistral-v3-tekken, mistral-v7, mistral-v7-tekken, monarch, openchat, orion, phi3, phi4, rwkv-world, seed_oss, smolvlm, vicuna, vicuna-orca, yandex, zephyr<br/>(env: LLAMA_ARG_CHAT_TEMPLATE) |
 | `--chat-template-file JINJA_TEMPLATE_FILE` | set custom jinja chat template file (default: template taken from model's metadata)<br/>if suffix/prefix are specified, template will be disabled<br/>only commonly used templates are accepted (unless --jinja is set before this flag):<br/>list of built-in templates:<br/>bailing, chatglm3, chatglm4, chatml, command-r, deepseek, deepseek2, deepseek3, exaone3, exaone4, falcon3, gemma, gigachat, glmedge, gpt-oss, granite, hunyuan-dense, hunyuan-moe, kimi-k2, llama2, llama2-sys, llama2-sys-bos, llama2-sys-strip, llama3, llama4, megrez, minicpm, mistral-v1, mistral-v3, mistral-v3-tekken, mistral-v7, mistral-v7-tekken, monarch, openchat, orion, phi3, phi4, rwkv-world, seed_oss, smolvlm, vicuna, vicuna-orca, yandex, zephyr<br/>(env: LLAMA_ARG_CHAT_TEMPLATE_FILE) |

From 091564699e38f4d11173d9dd10d5667791f6ea30 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Fri, 3 Oct 2025 16:22:47 +0200
Subject: [PATCH 03/10] refactor: address review feedback from allozaur

- Passed the assistant message content directly to ChatMessageAssistant to drop the redundant derived state in the chat message component
- Simplified chat streaming updates by removing unused partial-thinking handling and persisting partial responses straight from currentResponse
- Refreshed the ChatMessage stories to cover standard and reasoning scenarios without the old THINK-tag parsing examples

Co-authored-by: Aleksander Grygier <aleksander.grygier@gmail.com>
---
 .../app/chat/ChatMessages/ChatMessage.svelte  |   4 +-
 .../webui/src/lib/stores/chat.svelte.ts       |   7 +-
 .../src/stories/ChatMessage.stories.svelte    | 187 +++---------------
 3 files changed, 27 insertions(+), 171 deletions(-)

diff --git a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
index 83bde07abf6da..27bbf0a02b13d 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
@@ -52,8 +52,6 @@
 		return null;
 	});
 
-	let messageContent = $derived.by(() => message.content);
-
 	function handleCancelEdit() {
 		isEditing = false;
 		editedContent = message.content;
@@ -152,7 +150,7 @@
 		{editedContent}
 		{isEditing}
 		{message}
-		{messageContent}
+		messageContent={message.content}
 		onCancelEdit={handleCancelEdit}
 		onConfirmDelete={handleConfirmDelete}
 		onCopy={handleCopy}
diff --git a/tools/server/webui/src/lib/stores/chat.svelte.ts b/tools/server/webui/src/lib/stores/chat.svelte.ts
index e839d8693c683..f1377e8329f64 100644
--- a/tools/server/webui/src/lib/stores/chat.svelte.ts
+++ b/tools/server/webui/src/lib/stores/chat.svelte.ts
@@ -340,6 +340,7 @@ class ChatStore {
 				streamedContent += chunk;
 				this.currentResponse = streamedContent;
 
+				captureModelIfNeeded();
 				const messageIndex = this.findMessageIndex(assistantMessage.id);
 				this.updateMessageAtIndex(messageIndex, {
 					content: streamedContent
@@ -690,14 +691,12 @@ class ChatStore {
 
 		if (lastMessage && lastMessage.role === 'assistant') {
 			try {
-				const contentToSave = this.currentResponse;
-
 				const updateData: {
 					content: string;
 					thinking?: string;
 					timings?: ChatMessageTimings;
 				} = {
-					content: contentToSave
+					content: this.currentResponse
 				};
 
 				if (lastMessage.thinking?.trim()) {
@@ -721,7 +720,7 @@ class ChatStore {
 
 				await DatabaseStore.updateMessage(lastMessage.id, updateData);
 
-				lastMessage.content = contentToSave;
+				lastMessage.content = this.currentResponse;
 				if (updateData.thinking !== undefined) {
 					lastMessage.thinking = updateData.thinking;
 				}
diff --git a/tools/server/webui/src/stories/ChatMessage.stories.svelte b/tools/server/webui/src/stories/ChatMessage.stories.svelte
index e35b3618586d0..5d768c0854ce7 100644
--- a/tools/server/webui/src/stories/ChatMessage.stories.svelte
+++ b/tools/server/webui/src/stories/ChatMessage.stories.svelte
@@ -36,66 +36,25 @@
 		children: []
 	};
 
-	let processingMessage = $state({
-		id: '4',
-		convId: 'conv-1',
-		type: 'message',
-		timestamp: 0, // No timestamp = processing
-		role: 'assistant',
-		content: '',
-		parent: '1',
-		thinking: '',
-		children: []
-	});
-
-	let streamingMessage = $state({
-		id: '5',
-		convId: 'conv-1',
-		type: 'message',
-		timestamp: 0, // No timestamp = streaming
-		role: 'assistant',
-		content: '',
-		parent: '1',
-		thinking: '',
-		children: []
-	});
-
-	// Message with dedicated thinking content
-	const thinkTagMessage: DatabaseMessage = {
-		id: '6',
+	const assistantWithReasoning: DatabaseMessage = {
+		id: '3',
 		convId: 'conv-1',
 		type: 'message',
 		timestamp: Date.now() - 1000 * 60 * 2,
 		role: 'assistant',
 		content:
-			"Here's my response after thinking through the problem. The reasoning should appear separately from this main response content.",
+			"Here's the concise answer, now that I've thought it through carefully for you.",
 		parent: '1',
 		thinking:
-			"Let me analyze this step by step:\n\n1. The user is asking about thinking formats\n2. I need to demonstrate the &lt;think&gt; tag format\n3. This content should be displayed in the thinking section\n4. The main response should be separate\n\nThis is a good example of reasoning content.",
+			"Let's consider the user's question step by step:\\n\\n1. Identify the core problem\\n2. Evaluate relevant information\\n3. Formulate a clear answer\\n\\nFollowing this process ensures the final response stays focused and accurate.",
 		children: []
 	};
 
-	// Message with separate DeepSeek-style thinking content
-	const thinkBracketMessage: DatabaseMessage = {
-		id: '7',
-		convId: 'conv-1',
-		type: 'message',
-		timestamp: Date.now() - 1000 * 60 * 1,
-		role: 'assistant',
-		content:
-			"Here's my response after using the [THINK] format to organize the answer. The reasoning content should be shown separately from this response.",
-		parent: '1',
-		thinking:
-			"This is the DeepSeek-style thinking format:\n\n- Using square brackets instead of angle brackets\n- Should work identically to the &lt;think&gt; format\n- Reasoning content should display in the thinking section\n- Main response should be separate\n\nBoth formats should be supported seamlessly.",
-		children: []
-	};
-
-	// Streaming message for dedicated thinking content
-	let streamingThinkMessage = $state({
-		id: '8',
+	let processingMessage = $state({
+		id: '4',
 		convId: 'conv-1',
 		type: 'message',
-		timestamp: 0, // No timestamp = streaming
+		timestamp: 0, // No timestamp = processing
 		role: 'assistant',
 		content: '',
 		parent: '1',
@@ -103,9 +62,8 @@
 		children: []
 	});
 
-		// Streaming message for DeepSeek-style thinking content
-	let streamingBracketMessage = $state({
-		id: '9',
+	let streamingMessage = $state({
+		id: '5',
 		convId: 'conv-1',
 		type: 'message',
 		timestamp: 0, // No timestamp = streaming
@@ -133,11 +91,19 @@
 />
 
 <Story
-	name="WithThinkingBlock"
+	name="AssistantWithReasoning"
+	args={{
+		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
+		message: assistantWithReasoning
+	}}
+/>
+
+<Story
+	name="WithReasoningContent"
 	args={{
 		message: streamingMessage
 	}}
-	asChild
+asChild
 	play={async () => {
 		// Phase 1: Stream reasoning content in chunks
 		let reasoningText =
@@ -183,6 +149,7 @@
 	</div>
 </Story>
 
+
 <Story
 	name="Processing"
 	args={{
@@ -191,120 +158,12 @@
 	play={async () => {
 		// Import the chat store to simulate loading state
 		const { chatStore } = await import('$lib/stores/chat.svelte');
-		
+
 		// Set loading state to true to trigger the processing UI
 		chatStore.isLoading = true;
-		
+
 		// Simulate the processing state hook behavior
 		// This will show the "Generating..." text and parameter details
-		await new Promise(resolve => setTimeout(resolve, 100));
-	}}
-/>
-
-<Story
-	name="ThinkTagFormat"
-	args={{
-		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
-		message: thinkTagMessage
+		await new Promise((resolve) => setTimeout(resolve, 100));
 	}}
 />
-
-<Story
-	name="ThinkBracketFormat"
-	args={{
-		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
-		message: thinkBracketMessage
-	}}
-/>
-
-<Story
-	name="StreamingThinkTag"
-	args={{
-		message: streamingThinkMessage
-	}}
-	parameters={{
-		test: {
-			timeout: 30000
-		}
-	}}
-	asChild
-	play={async () => {
-		// Phase 1: Stream reasoning content
-		const thinkingContent =
-			'Let me work through this problem systematically:\n\n1. First, I need to understand what the user is asking\n2. Then I should consider different approaches\n3. I need to evaluate the pros and cons\n4. Finally, I should provide a clear recommendation\n\nThis step-by-step approach will ensure accuracy.';
-
-		streamingThinkMessage.thinking = '';
-		streamingThinkMessage.content = '';
-
-		for (let i = 0; i < thinkingContent.length; i++) {
-			streamingThinkMessage.thinking += thinkingContent[i];
-			await new Promise((resolve) => setTimeout(resolve, 5));
-		}
-
-		await new Promise((resolve) => setTimeout(resolve, 200));
-
-		// Phase 2: Stream main response content
-		const responseContent =
-			"Based on my analysis above, here's the solution:\n\n**Key Points:**\n- The approach should be systematic\n- We need to consider all factors\n- Implementation should be step-by-step\n\nThis ensures the best possible outcome.";
-
-		let currentContent = '';
-
-		for (let i = 0; i < responseContent.length; i++) {
-			currentContent += responseContent[i];
-			streamingThinkMessage.content = currentContent;
-			await new Promise((resolve) => setTimeout(resolve, 10));
-		}
-
-		streamingThinkMessage.timestamp = Date.now();
-	}}
->
-	<div class="w-[56rem]">
-		<ChatMessage message={streamingThinkMessage} />
-	</div>
-</Story>
-
-<Story
-	name="StreamingThinkBracket"
-	args={{
-		message: streamingBracketMessage
-	}}
-	parameters={{
-		test: {
-			timeout: 30000
-		}
-	}}
-	asChild
-	play={async () => {
-		// Phase 1: Stream DeepSeek-style reasoning content
-		const thinkingContent =
-			'Using the DeepSeek format now:\n\n- This demonstrates the &#91;THINK&#93; bracket format\n- Should behave identically to the &lt;think&gt; format\n- The UI should display this in the thinking section\n- Main content should be separate\n\nBoth formats provide the same functionality.';
-
-		streamingBracketMessage.thinking = '';
-		streamingBracketMessage.content = '';
-
-		for (let i = 0; i < thinkingContent.length; i++) {
-			streamingBracketMessage.thinking += thinkingContent[i];
-			await new Promise((resolve) => setTimeout(resolve, 5));
-		}
-
-		await new Promise((resolve) => setTimeout(resolve, 200));
-
-		// Phase 2: Stream main response content
-		const responseContent =
-			"Here's my response after using the &#91;THINK&#93; format:\n\n**Observations:**\n- Both &lt;think&gt; and &#91;THINK&#93; formats work seamlessly\n- The helper logic handles both cases\n- UI display is consistent across formats\n\nThis demonstrates the enhanced thinking content support.";
-
-		let currentContent = '';
-
-		for (let i = 0; i < responseContent.length; i++) {
-			currentContent += responseContent[i];
-			streamingBracketMessage.content = currentContent;
-			await new Promise((resolve) => setTimeout(resolve, 10));
-		}
-
-		streamingBracketMessage.timestamp = Date.now();
-	}}
->
-	<div class="w-[56rem]">
-		<ChatMessage message={streamingBracketMessage} />
-	</div>
-</Story>

From 77cce204d47095b78272ea36a4276096063026b6 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Fri, 3 Oct 2025 18:29:47 +0200
Subject: [PATCH 04/10] refactor: restore forced reasoning prefix to pass
 test-chat ([chat] All tests passed)

- store the exact sequence seen on input when 'thinking_forced_open' enforces a reasoning block
- inject this prefix before the first accumulated segment in 'reasoning_content', then clear it to avoid duplication
- repeat the capture on every new 'start_think' detection to properly handle partial/streaming flows
---
 common/chat-parser.cpp | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/common/chat-parser.cpp b/common/chat-parser.cpp
index 392f0f587b9f1..ea54b29b1c174 100644
--- a/common/chat-parser.cpp
+++ b/common/chat-parser.cpp
@@ -169,6 +169,23 @@ void common_chat_msg_parser::consume_literal(const std::string & literal) {
 }
 
 bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think, const std::string & end_think) {
+    std::string pending_reasoning_prefix;
+
+    auto set_reasoning_prefix = [&](size_t prefix_pos) {
+        if (!syntax_.thinking_forced_open || syntax_.reasoning_in_content) {
+            return;
+        }
+        if (prefix_pos + start_think.size() > input_.size()) {
+            pending_reasoning_prefix.clear();
+            return;
+        }
+        // Capture the exact literal that opened the reasoning section so we can
+        // surface it back to callers. This ensures formats that force the
+        // reasoning tag open (e.g. DeepSeek R1) retain their original prefix
+        // instead of dropping it during parsing.
+        pending_reasoning_prefix = input_.substr(prefix_pos, start_think.size());
+    };
+
     auto handle_reasoning = [&](const std::string & reasoning, bool closed) {
         auto stripped_reasoning = string_strip(reasoning);
         if (stripped_reasoning.empty()) {
@@ -181,6 +198,10 @@ bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think
                 add_content(syntax_.reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK ? "</think>" : end_think);
             }
         } else {
+            if (!pending_reasoning_prefix.empty()) {
+                add_reasoning_content(pending_reasoning_prefix);
+                pending_reasoning_prefix.clear();
+            }
             add_reasoning_content(stripped_reasoning);
         }
     };
@@ -233,6 +254,7 @@ bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think
         if (whitespace_end > pos_) {
             add_content(input_.substr(pos_, whitespace_end - pos_));
         }
+        set_reasoning_prefix(cursor);
         cursor += start_think.size();
     } else if (syntax_.thinking_forced_open) {
         cursor = whitespace_end;
@@ -282,6 +304,7 @@ bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think
                 move_to(input_.size());
                 return true;
             }
+            set_reasoning_prefix(cursor);
             cursor += start_think.size();
             continue;
         }

From a2fdf42273ce1a2a10928119e760106e2b0e5690 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Fri, 3 Oct 2025 20:59:03 +0200
Subject: [PATCH 05/10] refactor: address review feedback from ngxson

---
 common/arg.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/common/arg.cpp b/common/arg.cpp
index 53fc4e39b3654..ac27d98408c10 100644
--- a/common/arg.cpp
+++ b/common/arg.cpp
@@ -3427,7 +3427,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
         "- none: leaves thoughts unparsed in `message.content`\n"
         "- deepseek: puts thoughts in `message.reasoning_content`\n"
         "- deepseek-legacy: keeps `<think>` tags in `message.content` while also populating `message.reasoning_content`\n"
-        "(default: deepseek)",
+        "(default: auto)",
         [](common_params & params, const std::string & value) {
             params.reasoning_format = common_reasoning_format_from_name(value);
         }

From 3557493c5bc01def4df5272f65f5857c7bbc9868 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Sat, 4 Oct 2025 08:14:36 +0200
Subject: [PATCH 06/10] debug: say goodbye to curl -N, hello one-click raw
 stream

- adds a new checkbox in the WebUI to display raw LLM output without backend parsing or frontend Markdown rendering
---
 .../ChatMessages/ChatMessageAssistant.svelte  | 24 ++++++++++++++++++-
 .../ChatSettings/ChatSettingsDialog.svelte    |  5 ++++
 .../src/lib/constants/settings-config.ts      |  3 +++
 tools/server/webui/src/lib/services/chat.ts   |  5 ++--
 4 files changed, 34 insertions(+), 3 deletions(-)

diff --git a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
index ad3ffa3792d58..2f536b2c79c21 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
@@ -131,7 +131,11 @@
 			</div>
 		</div>
 	{:else if message.role === 'assistant'}
-		<MarkdownContent content={messageContent || ''} />
+		{#if config().disableReasoningFormat}
+			<pre class="raw-output">{messageContent || ''}</pre>
+		{:else}
+			<MarkdownContent content={messageContent || ''} />
+		{/if}
 	{:else}
 		<div class="text-sm whitespace-pre-wrap">
 			{messageContent}
@@ -203,4 +207,22 @@
 			background-position: -200% 0;
 		}
 	}
+
+	.raw-output {
+		width: 100%;
+		max-width: 48rem;
+		margin-top: 1.5rem;
+		padding: 1rem 1.25rem;
+		border-radius: 1rem;
+		background: hsl(var(--muted) / 0.3);
+		color: var(--foreground);
+		font-family:
+			ui-monospace, SFMono-Regular, 'SF Mono', Monaco, 'Cascadia Code', 'Roboto Mono', Consolas,
+			'Liberation Mono', Menlo, monospace;
+		font-size: 0.875rem;
+		line-height: 1.6;
+		white-space: pre-wrap;
+		word-break: break-word;
+	}
+
 </style>
diff --git a/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte b/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
index d832abc2e2831..185b76c0cc0cb 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
@@ -148,6 +148,11 @@
 					key: 'showThoughtInProgress',
 					label: 'Show thought in progress',
 					type: 'checkbox'
+				},
+				{
+					key: 'disableReasoningFormat',
+					label: 'Show raw LLM output without backend parsing and frontend Markdown rendering to inspect streaming across different models.',
+					type: 'checkbox'
 				}
 			]
 		},
diff --git a/tools/server/webui/src/lib/constants/settings-config.ts b/tools/server/webui/src/lib/constants/settings-config.ts
index 5cb22f4c34082..63e4364ae5656 100644
--- a/tools/server/webui/src/lib/constants/settings-config.ts
+++ b/tools/server/webui/src/lib/constants/settings-config.ts
@@ -6,6 +6,7 @@ export const SETTING_CONFIG_DEFAULT: Record<string, string | number | boolean> =
 	theme: 'system',
 	showTokensPerSecond: false,
 	showThoughtInProgress: false,
+	disableReasoningFormat: false,
 	keepStatsVisible: false,
 	askForTitleConfirmation: false,
 	pasteLongTextToFileLen: 2500,
@@ -76,6 +77,8 @@ export const SETTING_CONFIG_INFO: Record<string, string> = {
 	custom: 'Custom JSON parameters to send to the API. Must be valid JSON format.',
 	showTokensPerSecond: 'Display generation speed in tokens per second during streaming.',
 	showThoughtInProgress: 'Expand thought process by default when generating messages.',
+	disableReasoningFormat:
+		'Show raw LLM output without backend parsing and frontend Markdown rendering to inspect streaming across different models.',
 	keepStatsVisible: 'Keep processing statistics visible after generation finishes.',
 	askForTitleConfirmation:
 		'Ask for confirmation before automatically changing conversation title when editing the first message.',
diff --git a/tools/server/webui/src/lib/services/chat.ts b/tools/server/webui/src/lib/services/chat.ts
index 90df8fa730ba1..2e3a8d43b51cf 100644
--- a/tools/server/webui/src/lib/services/chat.ts
+++ b/tools/server/webui/src/lib/services/chat.ts
@@ -78,6 +78,8 @@ export class ChatService {
 			timings_per_token
 		} = options;
 
+		const currentConfig = config();
+
 		// Cancel any ongoing request and create a new abort controller
 		this.abort();
 		this.abortController = new AbortController();
@@ -117,7 +119,7 @@ export class ChatService {
 			stream
 		};
 
-		requestBody.reasoning_format = 'auto';
+		requestBody.reasoning_format = currentConfig.disableReasoningFormat ? 'none' : 'auto';
 
 		if (temperature !== undefined) requestBody.temperature = temperature;
 		// Set max_tokens to -1 (infinite) if not provided or empty
@@ -161,7 +163,6 @@ export class ChatService {
 		}
 
 		try {
-			const currentConfig = config();
 			const apiKey = currentConfig.apiKey?.toString().trim();
 
 			const response = await fetch(`./v1/chat/completions`, {

From 1e6beb64054a20a9b336ff04ce7e4bf9036ac1c3 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Mon, 6 Oct 2025 13:30:45 +0200
Subject: [PATCH 07/10] Update
 tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte

Co-authored-by: Aleksander Grygier <aleksander.grygier@gmail.com>
---
 .../src/lib/components/app/chat/ChatMessages/ChatMessage.svelte  | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
index 27bbf0a02b13d..fed0cf712695f 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte
@@ -47,6 +47,7 @@
 	let thinkingContent = $derived.by(() => {
 		if (message.role === 'assistant') {
 			const trimmedThinking = message.thinking?.trim();
+
 			return trimmedThinking ? trimmedThinking : null;
 		}
 		return null;

From 5e1604477a62c965f7da8cf389b397b07baa4aa8 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Mon, 6 Oct 2025 14:33:23 +0200
Subject: [PATCH 08/10] webui: add Storybook example for raw LLM output and
 scope reasoning format toggle per story

- Added a Storybook example that showcases the chat message component in raw LLM output mode with the provided trace sample
- Updated every ChatMessage story to toggle the disableReasoningFormat setting so the raw-output rendering remains scoped to its own example
---
 .../src/stories/ChatMessage.stories.svelte    | 41 +++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/tools/server/webui/src/stories/ChatMessage.stories.svelte b/tools/server/webui/src/stories/ChatMessage.stories.svelte
index 5d768c0854ce7..634ff22799107 100644
--- a/tools/server/webui/src/stories/ChatMessage.stories.svelte
+++ b/tools/server/webui/src/stories/ChatMessage.stories.svelte
@@ -49,6 +49,19 @@
 			"Let's consider the user's question step by step:\\n\\n1. Identify the core problem\\n2. Evaluate relevant information\\n3. Formulate a clear answer\\n\\nFollowing this process ensures the final response stays focused and accurate.",
 		children: []
 	};
+	const rawOutputMessage: DatabaseMessage = {
+		id: '6',
+		convId: 'conv-1',
+		type: 'message',
+		timestamp: Date.now() - 1000 * 60,
+		role: 'assistant',
+		content:
+			'<|channel|>analysis<|message|>User greeted me. Initiating overcomplicated analysis: Is this a trap? No, just a normal hello. Respond calmly, act like a helpful assistant, and do not start explaining quantum physics again. Confidence 0.73. Engaging socially acceptable greeting protocol...<|end|>Hello there! How can I help you today?',
+		parent: '1',
+		thinking: '',
+		children: []
+	};
+
 
 	let processingMessage = $state({
 		id: '4',
@@ -80,6 +93,10 @@
 	args={{
 		message: userMessage
 	}}
+	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', false);
+	}}
 />
 
 <Story
@@ -88,6 +105,10 @@
 		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
 		message: assistantMessage
 	}}
+	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', false);
+	}}
 />
 
 <Story
@@ -96,6 +117,22 @@
 		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
 		message: assistantWithReasoning
 	}}
+	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', false);
+	}}
+/>
+
+<Story
+	name="RawLlmOutput"
+	args={{
+		class: 'max-w-[56rem] w-[calc(100vw-2rem)]',
+		message: rawOutputMessage
+	}}
+	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', true);
+	}}
 />
 
 <Story
@@ -105,6 +142,8 @@
 	}}
 asChild
 	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', false);
 		// Phase 1: Stream reasoning content in chunks
 		let reasoningText =
 			'I need to think about this carefully. Let me break down the problem:\n\n1. The user is asking for help with something complex\n2. I should provide a thorough and helpful response\n3. I need to consider multiple approaches\n4. The best solution would be to explain step by step\n\nThis approach will ensure clarity and understanding.';
@@ -156,6 +195,8 @@ asChild
 		message: processingMessage
 	}}
 	play={async () => {
+		const { updateConfig } = await import('$lib/stores/settings.svelte');
+		updateConfig('disableReasoningFormat', false);
 		// Import the chat store to simulate loading state
 		const { chatStore } = await import('$lib/stores/chat.svelte');
 

From 73ae73b6de3ab37f31c1e126aa0ac541cb6d6731 Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Mon, 6 Oct 2025 16:13:42 +0200
Subject: [PATCH 09/10] npm run format

---
 .../ChatMessages/ChatMessageAssistant.svelte  |  1 -
 .../ChatSettings/ChatSettingsDialog.svelte    |  3 ++-
 tools/server/webui/src/lib/services/chat.ts   | 22 +++++++++----------
 .../src/stories/ChatMessage.stories.svelte    |  7 ++----
 4 files changed, 15 insertions(+), 18 deletions(-)

diff --git a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
index 2f536b2c79c21..5539ed9e21c34 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte
@@ -224,5 +224,4 @@
 		white-space: pre-wrap;
 		word-break: break-word;
 	}
-
 </style>
diff --git a/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte b/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
index 185b76c0cc0cb..7c25e59258829 100644
--- a/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
+++ b/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsDialog.svelte
@@ -151,7 +151,8 @@
 				},
 				{
 					key: 'disableReasoningFormat',
-					label: 'Show raw LLM output without backend parsing and frontend Markdown rendering to inspect streaming across different models.',
+					label:
+						'Show raw LLM output without backend parsing and frontend Markdown rendering to inspect streaming across different models.',
 					type: 'checkbox'
 				}
 			]
diff --git a/tools/server/webui/src/lib/services/chat.ts b/tools/server/webui/src/lib/services/chat.ts
index 2e3a8d43b51cf..e1dad24a8f246 100644
--- a/tools/server/webui/src/lib/services/chat.ts
+++ b/tools/server/webui/src/lib/services/chat.ts
@@ -307,17 +307,17 @@ export class ChatService {
 								}
 							}
 
-                                                        if (content) {
-                                                                hasReceivedData = true;
-                                                                aggregatedContent += content;
-                                                                onChunk?.(content);
-                                                        }
-
-                                                        if (reasoningContent) {
-                                                                hasReceivedData = true;
-                                                                fullReasoningContent += reasoningContent;
-                                                                onReasoningChunk?.(reasoningContent);
-                                                        }
+							if (content) {
+								hasReceivedData = true;
+								aggregatedContent += content;
+								onChunk?.(content);
+							}
+
+							if (reasoningContent) {
+								hasReceivedData = true;
+								fullReasoningContent += reasoningContent;
+								onReasoningChunk?.(reasoningContent);
+							}
 						} catch (e) {
 							console.error('Error parsing JSON chunk:', e);
 						}
diff --git a/tools/server/webui/src/stories/ChatMessage.stories.svelte b/tools/server/webui/src/stories/ChatMessage.stories.svelte
index 634ff22799107..6529b75a307cd 100644
--- a/tools/server/webui/src/stories/ChatMessage.stories.svelte
+++ b/tools/server/webui/src/stories/ChatMessage.stories.svelte
@@ -42,8 +42,7 @@
 		type: 'message',
 		timestamp: Date.now() - 1000 * 60 * 2,
 		role: 'assistant',
-		content:
-			"Here's the concise answer, now that I've thought it through carefully for you.",
+		content: "Here's the concise answer, now that I've thought it through carefully for you.",
 		parent: '1',
 		thinking:
 			"Let's consider the user's question step by step:\\n\\n1. Identify the core problem\\n2. Evaluate relevant information\\n3. Formulate a clear answer\\n\\nFollowing this process ensures the final response stays focused and accurate.",
@@ -62,7 +61,6 @@
 		children: []
 	};
 
-
 	let processingMessage = $state({
 		id: '4',
 		convId: 'conv-1',
@@ -140,7 +138,7 @@
 	args={{
 		message: streamingMessage
 	}}
-asChild
+	asChild
 	play={async () => {
 		const { updateConfig } = await import('$lib/stores/settings.svelte');
 		updateConfig('disableReasoningFormat', false);
@@ -188,7 +186,6 @@ asChild
 	</div>
 </Story>
 
-
 <Story
 	name="Processing"
 	args={{

From 9074d040755192b37b94e19dbcbc36b45f791bed Mon Sep 17 00:00:00 2001
From: Pascal <admin@serveurperso.com>
Date: Mon, 6 Oct 2025 20:05:55 +0200
Subject: [PATCH 10/10] chat-parser: address review feedback from ngxson

Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
---
 common/chat-parser.cpp | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/common/chat-parser.cpp b/common/chat-parser.cpp
index ea54b29b1c174..7365782e7d6d8 100644
--- a/common/chat-parser.cpp
+++ b/common/chat-parser.cpp
@@ -171,6 +171,10 @@ void common_chat_msg_parser::consume_literal(const std::string & literal) {
 bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think, const std::string & end_think) {
     std::string pending_reasoning_prefix;
 
+    if (syntax_.reasoning_format == COMMON_REASONING_FORMAT_NONE) {
+        return false;
+    }
+
     auto set_reasoning_prefix = [&](size_t prefix_pos) {
         if (!syntax_.thinking_forced_open || syntax_.reasoning_in_content) {
             return;
@@ -206,10 +210,6 @@ bool common_chat_msg_parser::try_parse_reasoning(const std::string & start_think
         }
     };
 
-    if (syntax_.reasoning_format == COMMON_REASONING_FORMAT_NONE) {
-        return false;
-    }
-
     const size_t saved_pos = pos_;
     const size_t saved_content_size = result_.content.size();
     const size_t saved_reasoning_size = result_.reasoning_content.size();