merged new updates

CHANGELOG.md

@@ -5,6 +5,26 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.108] - 2024-03-02
+
+### Added
+
+- **🎮 Playground Feature (Beta)**: Explore the full potential of the raw API through an intuitive UI with our new playground feature, accessible to admins. Simply click on the bottom name area of the sidebar to access it. The playground feature offers two modes text completion (notebook) and chat completion. As it's in beta, please report any issues you encounter.
+- **🛠️ Direct Database Download for Admins**: Admins can now download the database directly from the WebUI via the admin settings.
+- **🎨 Additional RAG Settings**: Customize your RAG process with the ability to edit the TOP K value. Navigate to Documents > Settings > General to make changes.
+- **🖥️ UI Improvements**: Tooltips now available in the input area and sidebar handle. More tooltips will be added across other parts of the UI.
+
+### Fixed
+
+- Resolved input autofocus issue on mobile when the sidebar is open, making it easier to use.
+- Corrected numbered list display issue in Safari (#963).
+- Restricted user ability to delete chats without proper permissions (#993).
+
+### Changed
+
+- **Simplified Ollama Settings**: Ollama settings now don't require the `/api` suffix. You can now utilize the Ollama base URL directly, e.g., `http://localhost:11434`. Also, an `OLLAMA_BASE_URL` environment variable has been added.
+- **Database Renaming**: Starting from this release, `ollama.db` will be automatically renamed to `webui.db`.
+
 ## [0.1.107] - 2024-03-01
 
 ### Added

backend/apps/ollama/main.py

@@ -11,7 +11,7 @@ from pydantic import BaseModel
 from apps.web.models.users import Users
 from constants import ERROR_MESSAGES
 from utils.utils import decode_token, get_current_user, get_admin_user
-from config import OLLAMA_API_BASE_URL, WEBUI_AUTH
+from config import OLLAMA_BASE_URL, WEBUI_AUTH
 
 app = FastAPI()
 app.add_middleware(
@@ -22,7 +22,7 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
-app.state.OLLAMA_API_BASE_URL = OLLAMA_API_BASE_URL
+app.state.OLLAMA_BASE_URL = OLLAMA_BASE_URL
 
 # TARGET_SERVER_URL = OLLAMA_API_BASE_URL
 
@@ -32,7 +32,7 @@ REQUEST_POOL = []
 
 @app.get("/url")
 async def get_ollama_api_url(user=Depends(get_admin_user)):
-    return {"OLLAMA_API_BASE_URL": app.state.OLLAMA_API_BASE_URL}
+    return {"OLLAMA_BASE_URL": app.state.OLLAMA_BASE_URL}
 
 
 class UrlUpdateForm(BaseModel):
@@ -41,8 +41,8 @@ class UrlUpdateForm(BaseModel):
 
 @app.post("/url/update")
 async def update_ollama_api_url(form_data: UrlUpdateForm, user=Depends(get_admin_user)):
-    app.state.OLLAMA_API_BASE_URL = form_data.url
+    app.state.OLLAMA_BASE_URL = form_data.url
-    return {"OLLAMA_API_BASE_URL": app.state.OLLAMA_API_BASE_URL}
+    return {"OLLAMA_BASE_URL": app.state.OLLAMA_BASE_URL}
 
 
 @app.get("/cancel/{request_id}")
@@ -57,7 +57,7 @@ async def cancel_ollama_request(request_id: str, user=Depends(get_current_user))
 
 @app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
 async def proxy(path: str, request: Request, user=Depends(get_current_user)):
-    target_url = f"{app.state.OLLAMA_API_BASE_URL}/{path}"
+    target_url = f"{app.state.OLLAMA_BASE_URL}/{path}"
 
     body = await request.body()
     headers = dict(request.headers)
@@ -91,7 +91,13 @@ async def proxy(path: str, request: Request, user=Depends(get_current_user)):
 
             def stream_content():
                 try:
-                    if path in ["chat"]:
+                    if path == "generate":
+                        data = json.loads(body.decode("utf-8"))
+
+                        if not ("stream" in data and data["stream"] == False):
+                            yield json.dumps({"id": request_id, "done": False}) + "\n"
+
+                    elif path == "chat":
                         yield json.dumps({"id": request_id, "done": False}) + "\n"
 
                     for chunk in r.iter_content(chunk_size=8192):
@@ -103,6 +109,7 @@ async def proxy(path: str, request: Request, user=Depends(get_current_user)):
                 finally:
                     if hasattr(r, "close"):
                         r.close()
+                        if request_id in REQUEST_POOL:
                             REQUEST_POOL.remove(request_id)
 
             r = requests.request(

backend/apps/rag/main.py

@@ -79,6 +79,8 @@ app.state.CHUNK_SIZE = CHUNK_SIZE
 app.state.CHUNK_OVERLAP = CHUNK_OVERLAP
 app.state.RAG_TEMPLATE = RAG_TEMPLATE
 app.state.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL
+app.state.TOP_K = 4
+
 app.state.sentence_transformer_ef = (
     embedding_functions.SentenceTransformerEmbeddingFunction(
         model_name=app.state.RAG_EMBEDDING_MODEL,
@@ -210,23 +212,33 @@ async def get_rag_template(user=Depends(get_current_user)):
     }
 
 
-class RAGTemplateForm(BaseModel):
+@app.get("/query/settings")
-    template: str
+async def get_query_settings(user=Depends(get_admin_user)):
+    return {
+        "status": True,
+        "template": app.state.RAG_TEMPLATE,
+        "k": app.state.TOP_K,
+    }
 
 
-@app.post("/template/update")
+class QuerySettingsForm(BaseModel):
-async def update_rag_template(form_data: RAGTemplateForm, user=Depends(get_admin_user)):
+    k: Optional[int] = None
-    # TODO: check template requirements
+    template: Optional[str] = None
-    app.state.RAG_TEMPLATE = (
+
-        form_data.template if form_data.template != "" else RAG_TEMPLATE
+
-    )
+@app.post("/query/settings/update")
+async def update_query_settings(
+    form_data: QuerySettingsForm, user=Depends(get_admin_user)
+):
+    app.state.RAG_TEMPLATE = form_data.template if form_data.template else RAG_TEMPLATE
+    app.state.TOP_K = form_data.k if form_data.k else 4
     return {"status": True, "template": app.state.RAG_TEMPLATE}
 
 
 class QueryDocForm(BaseModel):
     collection_name: str
     query: str
-    k: Optional[int] = 4
+    k: Optional[int] = None
 
 
 @app.post("/query/doc")
@@ -240,7 +252,10 @@ def query_doc(
             name=form_data.collection_name,
             embedding_function=app.state.sentence_transformer_ef,
         )
-        result = collection.query(query_texts=[form_data.query], n_results=form_data.k)
+        result = collection.query(
+            query_texts=[form_data.query],
+            n_results=form_data.k if form_data.k else app.state.TOP_K,
+        )
         return result
     except Exception as e:
         print(e)
@@ -253,7 +268,7 @@ def query_doc(
 class QueryCollectionsForm(BaseModel):
     collection_names: List[str]
     query: str
-    k: Optional[int] = 4
+    k: Optional[int] = None
 
 
 def merge_and_sort_query_results(query_results, k):
@@ -317,13 +332,16 @@ def query_collection(
             )
 
             result = collection.query(
-                query_texts=[form_data.query], n_results=form_data.k
+                query_texts=[form_data.query],
+                n_results=form_data.k if form_data.k else app.state.TOP_K,
             )
             results.append(result)
         except:
             pass
 
-    return merge_and_sort_query_results(results, form_data.k)
+    return merge_and_sort_query_results(
+        results, form_data.k if form_data.k else app.state.TOP_K
+    )
 
 
 @app.post("/web")
@@ -423,7 +441,9 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
         "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
     ] or file_ext in ["xls", "xlsx"]:
         loader = UnstructuredExcelLoader(file_path)
-    elif file_ext in known_source_ext or (file_content_type and file_content_type.find("text/") >= 0):
+    elif file_ext in known_source_ext or (
+        file_content_type and file_content_type.find("text/") >= 0
+    ):
         loader = TextLoader(file_path)
     else:
         loader = TextLoader(file_path)

backend/config.py

@@ -211,6 +211,17 @@ if ENV == "prod":
     if OLLAMA_API_BASE_URL == "/ollama/api":
         OLLAMA_API_BASE_URL = "http://host.docker.internal:11434/api"
 
+
+OLLAMA_BASE_URL = os.environ.get("OLLAMA_BASE_URL", "")
+
+if OLLAMA_BASE_URL == "":
+    OLLAMA_BASE_URL = (
+        OLLAMA_API_BASE_URL[:-4]
+        if OLLAMA_API_BASE_URL.endswith("/api")
+        else OLLAMA_API_BASE_URL
+    )
+
+
 ####################################
 # OPENAI_API
 ####################################

package.json

@@ -1,6 +1,6 @@
 {
 	"name": "open-webui",
-	"version": "0.1.107",
+	"version": "0.1.108",
 	"private": true,
 	"scripts": {
 		"dev": "vite dev --host",

src/lib/apis/ollama/index.ts

@@ -29,7 +29,7 @@ export const getOllamaAPIUrl = async (token: string = '') => {
 		throw error;
 	}
 
-	return res.OLLAMA_API_BASE_URL;
+	return res.OLLAMA_BASE_URL;
 };
 
 export const updateOllamaAPIUrl = async (token: string = '', url: string) => {
@@ -64,13 +64,13 @@ export const updateOllamaAPIUrl = async (token: string = '', url: string) => {
 		throw error;
 	}
 
-	return res.OLLAMA_API_BASE_URL;
+	return res.OLLAMA_BASE_URL;
 };
 
 export const getOllamaVersion = async (token: string = '') => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/version`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/version`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',
@@ -102,7 +102,7 @@ export const getOllamaVersion = async (token: string = '') => {
 export const getOllamaModels = async (token: string = '') => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/tags`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/tags`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',
@@ -148,7 +148,7 @@ export const generateTitle = async (
 
 	console.log(template);
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/generate`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/generate`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'text/event-stream',
@@ -186,7 +186,7 @@ export const generatePrompt = async (token: string = '', model: string, conversa
 		conversation = '[no existing conversation]';
 	}
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/generate`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/generate`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'text/event-stream',
@@ -217,11 +217,37 @@ export const generatePrompt = async (token: string = '', model: string, conversa
 	return res;
 };
 
+export const generateTextCompletion = async (token: string = '', model: string, text: string) => {
+	let error = null;
+
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/generate`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'text/event-stream',
+			Authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			model: model,
+			prompt: text,
+			stream: true
+		})
+	}).catch((err) => {
+		error = err;
+		return null;
+	});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const generateChatCompletion = async (token: string = '', body: object) => {
 	let controller = new AbortController();
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/chat`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/chat`, {
 		signal: controller.signal,
 		method: 'POST',
 		headers: {
@@ -265,7 +291,7 @@ export const cancelChatCompletion = async (token: string = '', requestId: string
 export const createModel = async (token: string, tagName: string, content: string) => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/create`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/create`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'text/event-stream',
@@ -290,7 +316,7 @@ export const createModel = async (token: string, tagName: string, content: strin
 export const deleteModel = async (token: string, tagName: string) => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/delete`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/delete`, {
 		method: 'DELETE',
 		headers: {
 			'Content-Type': 'text/event-stream',
@@ -324,7 +350,7 @@ export const deleteModel = async (token: string, tagName: string) => {
 export const pullModel = async (token: string, tagName: string) => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/pull`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/pull`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'text/event-stream',

src/lib/apis/rag/index.ts

@@ -85,17 +85,49 @@ export const getRAGTemplate = async (token: string) => {
 	return res?.template ?? '';
 };
 
-export const updateRAGTemplate = async (token: string, template: string) => {
+export const getQuerySettings = async (token: string) => {
 	let error = null;
 
-	const res = await fetch(`${RAG_API_BASE_URL}/template/update`, {
+	const res = await fetch(`${RAG_API_BASE_URL}/query/settings`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+type QuerySettings = {
+	k: number | null;
+	template: string | null;
+};
+
+export const updateQuerySettings = async (token: string, settings: QuerySettings) => {
+	let error = null;
+
+	const res = await fetch(`${RAG_API_BASE_URL}/query/settings/update`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'application/json',
 			Authorization: `Bearer ${token}`
 		},
 		body: JSON.stringify({
-			template: template
+			...settings
 		})
 	})
 		.then(async (res) => {
@@ -183,7 +215,7 @@ export const queryDoc = async (
 	token: string,
 	collection_name: string,
 	query: string,
-	k: number
+	k: number | null = null
 ) => {
 	let error = null;
 

src/lib/components/chat/MessageInput/Models.svelte

@@ -81,6 +81,9 @@
 								throw data;
 							}
 
+							if ('id' in data) {
+								console.log(data);
+							} else {
 								if (data.done == false) {
 									if (prompt == '' && data.response == '\n') {
 										continue;
@@ -93,6 +96,7 @@
 								}
 							}
 						}
+					}
 				} catch (error) {
 					console.log(error);
 					if ('detail' in error) {

src/lib/components/chat/Settings/Connections.svelte

@@ -121,7 +121,7 @@
 				<div class="flex-1 mr-2">
 					<input
 						class="w-full rounded py-2 px-4 text-sm dark:text-gray-300 dark:bg-gray-800 outline-none"
-						placeholder="Enter URL (e.g. http://localhost:11434/api)"
+						placeholder="Enter URL (e.g. http://localhost:11434)"
 						bind:value={API_BASE_URL}
 					/>
 				</div>

src/lib/components/documents/Settings/General.svelte

@@ -2,10 +2,10 @@
 	import { getDocs } from '$lib/apis/documents';
 	import {
 		getChunkParams,
-		getRAGTemplate,
+		getQuerySettings,
 		scanDocs,
 		updateChunkParams,
-		updateRAGTemplate
+		updateQuerySettings
 	} from '$lib/apis/rag';
 	import { documents } from '$lib/stores';
 	import { onMount, getContext } from 'svelte';
@@ -20,7 +20,10 @@
 	let chunkSize = 0;
 	let chunkOverlap = 0;
 
-	let template = '';
+	let querySettings = {
+		template: '',
+		k: 4
+	};
 
 	const scanHandler = async () => {
 		loading = true;
@@ -35,7 +38,7 @@
 
 	const submitHandler = async () => {
 		const res = await updateChunkParams(localStorage.token, chunkSize, chunkOverlap);
-		await updateRAGTemplate(localStorage.token, template);
+		querySettings = await updateQuerySettings(localStorage.token, querySettings);
 	};
 
 	onMount(async () => {
@@ -46,7 +49,7 @@
 			chunkOverlap = res.chunk_overlap;
 		}
 
-		template = await getRAGTemplate(localStorage.token);
+		querySettings = await getQuerySettings(localStorage.token);
 	});
 </script>
 
@@ -160,10 +163,44 @@
 				</div>
 			</div>
 
+			<div class=" text-sm font-medium">Query Params</div>
+
+			<div class=" flex">
+				<div class="  flex w-full justify-between">
+					<div class="self-center text-xs font-medium flex-1">Top K</div>
+
+					<div class="self-center p-3">
+						<input
+							class=" w-full rounded py-1.5 px-4 text-sm dark:text-gray-300 dark:bg-gray-800 outline-none border border-gray-100 dark:border-gray-600"
+							type="number"
+							placeholder="Enter Top K"
+							bind:value={querySettings.k}
+							autocomplete="off"
+							min="0"
+						/>
+					</div>
+				</div>
+
+				<!-- <div class="flex w-full">
+					<div class=" self-center text-xs font-medium min-w-fit">Chunk Overlap</div>
+
+					<div class="self-center p-3">
+						<input
+							class="w-full rounded py-1.5 px-4 text-sm dark:text-gray-300 dark:bg-gray-800 outline-none border border-gray-100 dark:border-gray-600"
+							type="number"
+							placeholder="Enter Chunk Overlap"
+							bind:value={chunkOverlap}
+							autocomplete="off"
+							min="0"
+						/>
+					</div>
+				</div> -->
+			</div>
+
 			<div>
 				<div class=" mb-2.5 text-sm font-medium">{$i18n.t('RAG Template')}</div>
 				<textarea
-					bind:value={template}
+					bind:value={querySettings.template}
 					class="w-full rounded p-4 text-sm dark:text-gray-300 dark:bg-gray-800 outline-none resize-none"
 					rows="4"
 				/>

src/lib/components/layout/Sidebar.svelte

@@ -38,7 +38,7 @@
 	let isEditing = false;
 
 	onMount(async () => {
-		if (window.innerWidth > 1280) {
+		if (window.innerWidth > 1024) {
 			show = true;
 		}
 		await chats.set(await getChatList(localStorage.token));
@@ -388,6 +388,11 @@
 									? 'bg-gray-900'
 									: ''} transition whitespace-nowrap text-ellipsis"
 								href="/c/{chat.id}"
+								on:click={() => {
+									if (window.innerWidth < 1024) {
+										show = false;
+									}
+								}}
 							>
 								<div class=" flex self-center flex-1 w-full">
 									<div
@@ -599,6 +604,32 @@
 										</div>
 										<div class=" self-center font-medium">{$i18n.t('Admin Panel')}</div>
 									</button>
+
+									<button
+										class="flex py-2.5 px-3.5 w-full hover:bg-gray-800 transition"
+										on:click={() => {
+											goto('/playground');
+											showDropdown = false;
+										}}
+									>
+										<div class=" self-center mr-3">
+											<svg
+												xmlns="http://www.w3.org/2000/svg"
+												fill="none"
+												viewBox="0 0 24 24"
+												stroke-width="1.5"
+												stroke="currentColor"
+												class="w-5 h-5"
+											>
+												<path
+													stroke-linecap="round"
+													stroke-linejoin="round"
+													d="m6.75 7.5 3 2.25-3 2.25m4.5 0h3m-9 8.25h13.5A2.25 2.25 0 0 0 21 18V6a2.25 2.25 0 0 0-2.25-2.25H5.25A2.25 2.25 0 0 0 3 6v12a2.25 2.25 0 0 0 2.25 2.25Z"
+												/>
+											</svg>
+										</div>
+										<div class=" self-center font-medium">Playground</div>
+									</button>
 								{/if}
 
 								<button

src/lib/components/playground/ChatCompletion.svelte

@@ -0,0 +1,105 @@
+<script lang="ts">
+	import { onMount } from 'svelte';
+
+	export let messages = [];
+
+	onMount(() => {
+		messages.forEach((message, idx) => {
+			let textareaElement = document.getElementById(`${message.role}-${idx}-textarea`);
+			textareaElement.style.height = '';
+			textareaElement.style.height = textareaElement.scrollHeight + 'px';
+		});
+	});
+</script>
+
+<div class="py-3 space-y-3">
+	{#each messages as message, idx}
+		<div class="flex gap-2 group">
+			<div class="flex items-start pt-1">
+				<button
+					class="px-2 py-1 text-sm font-semibold uppercase min-w-[6rem] text-left dark:group-hover:bg-gray-800 rounded-lg transition"
+					on:click={() => {
+						message.role = message.role === 'user' ? 'assistant' : 'user';
+					}}>{message.role}</button
+				>
+			</div>
+
+			<div class="flex-1">
+				<textarea
+					id="{message.role}-{idx}-textarea"
+					class="w-full bg-transparent outline-none rounded-lg p-2 text-sm resize-none overflow-hidden"
+					placeholder="Enter {message.role === 'user' ? 'a user' : 'an assistant'} message here"
+					rows="1"
+					on:input={(e) => {
+						e.target.style.height = '';
+						e.target.style.height = e.target.scrollHeight + 'px';
+					}}
+					on:focus={(e) => {
+						e.target.style.height = '';
+						e.target.style.height = e.target.scrollHeight + 'px';
+
+						// e.target.style.height = Math.min(e.target.scrollHeight, 200) + 'px';
+					}}
+					bind:value={message.content}
+				/>
+			</div>
+
+			<div class=" pt-1">
+				<button
+					class=" group-hover:text-gray-500 dark:text-gray-900 dark:hover:text-gray-300 transition"
+					on:click={() => {
+						messages = messages.filter((message, messageIdx) => messageIdx !== idx);
+					}}
+				>
+					<svg
+						xmlns="http://www.w3.org/2000/svg"
+						fill="none"
+						viewBox="0 0 24 24"
+						stroke-width="2"
+						stroke="currentColor"
+						class="w-5 h-5"
+					>
+						<path
+							stroke-linecap="round"
+							stroke-linejoin="round"
+							d="M15 12H9m12 0a9 9 0 1 1-18 0 9 9 0 0 1 18 0Z"
+						/>
+					</svg>
+				</button>
+			</div>
+		</div>
+
+		<hr class=" dark:border-gray-800" />
+	{/each}
+
+	<button
+		class="flex items-center gap-2 px-2 py-1"
+		on:click={() => {
+			console.log(messages.at(-1));
+			messages.push({
+				role: (messages.at(-1)?.role ?? 'assistant') === 'user' ? 'assistant' : 'user',
+				content: ''
+			});
+			messages = messages;
+		}}
+	>
+		<div>
+			<svg
+				xmlns="http://www.w3.org/2000/svg"
+				fill="none"
+				viewBox="0 0 24 24"
+				stroke-width="1.5"
+				stroke="currentColor"
+				class="w-5 h-5"
+			>
+				<path
+					stroke-linecap="round"
+					stroke-linejoin="round"
+					d="M12 9v6m3-3H9m12 0a9 9 0 1 1-18 0 9 9 0 0 1 18 0Z"
+				/>
+			</svg>
+		</div>
+
+		<div class=" text-sm font-medium">Add message</div>
+	</button>
+</div>

src/routes/(app)/+page.svelte

@@ -253,19 +253,17 @@
 			let relevantContexts = await Promise.all(
 				docs.map(async (doc) => {
 					if (doc.type === 'collection') {
-						return await queryCollection(localStorage.token, doc.collection_names, query, 4).catch(
+						return await queryCollection(localStorage.token, doc.collection_names, query).catch(
 							(error) => {
 								console.log(error);
 								return null;
 							}
 						);
 					} else {
-						return await queryDoc(localStorage.token, doc.collection_name, query, 4).catch(
+						return await queryDoc(localStorage.token, doc.collection_name, query).catch((error) => {
-							(error) => {
 							console.log(error);
 							return null;
-							}
+						});
-						);
 					}
 				})
 			);

src/routes/(app)/c/[id]/+page.svelte

@@ -263,19 +263,17 @@
 			let relevantContexts = await Promise.all(
 				docs.map(async (doc) => {
 					if (doc.type === 'collection') {
-						return await queryCollection(localStorage.token, doc.collection_names, query, 4).catch(
+						return await queryCollection(localStorage.token, doc.collection_names, query).catch(
 							(error) => {
 								console.log(error);
 								return null;
 							}
 						);
 					} else {
-						return await queryDoc(localStorage.token, doc.collection_name, query, 4).catch(
+						return await queryDoc(localStorage.token, doc.collection_name, query).catch((error) => {
-							(error) => {
 							console.log(error);
 							return null;
-							}
+						});
-						);
 					}
 				})
 			);

src/routes/(app)/playground/+page.svelte

@@ -0,0 +1,472 @@
+<script>
+	import { goto } from '$app/navigation';
+
+	import { onMount, tick } from 'svelte';
+
+	import { toast } from 'svelte-sonner';
+
+	import {
+		LITELLM_API_BASE_URL,
+		OLLAMA_API_BASE_URL,
+		OPENAI_API_BASE_URL,
+		WEBUI_API_BASE_URL
+	} from '$lib/constants';
+	import { WEBUI_NAME, config, user, models, settings } from '$lib/stores';
+
+	import { cancelChatCompletion, generateChatCompletion } from '$lib/apis/ollama';
+	import { generateOpenAIChatCompletion } from '$lib/apis/openai';
+
+	import { splitStream } from '$lib/utils';
+	import ChatCompletion from '$lib/components/playground/ChatCompletion.svelte';
+
+	let mode = 'chat';
+	let loaded = false;
+
+	let text = '';
+
+	let selectedModelId = '';
+
+	let loading = false;
+	let currentRequestId;
+	let stopResponseFlag = false;
+
+	let system = '';
+	let messages = [
+		{
+			role: 'user',
+			content: ''
+		}
+	];
+
+	const scrollToBottom = () => {
+		let element;
+
+		if (mode === 'chat') {
+			element = document.getElementById('messages-container');
+		} else {
+			element = document.getElementById('text-completion-textarea');
+		}
+
+		if (element) {
+			element.scrollTop = element?.scrollHeight;
+		}
+	};
+
+	// const cancelHandler = async () => {
+	// 	if (currentRequestId) {
+	// 		const res = await cancelChatCompletion(localStorage.token, currentRequestId);
+	// 		currentRequestId = null;
+	// 		loading = false;
+	// 	}
+	// };
+
+	const stopResponse = () => {
+		stopResponseFlag = true;
+		console.log('stopResponse');
+	};
+
+	const textCompletionHandler = async () => {
+		const model = $models.find((model) => model.id === selectedModelId);
+
+		const res = await generateOpenAIChatCompletion(
+			localStorage.token,
+			{
+				model: model.id,
+				stream: true,
+				messages: [
+					{
+						role: 'assistant',
+						content: text
+					}
+				]
+			},
+			model.external
+				? model.source === 'litellm'
+					? `${LITELLM_API_BASE_URL}/v1`
+					: `${OPENAI_API_BASE_URL}`
+				: `${OLLAMA_API_BASE_URL}/v1`
+		);
+
+		if (res && res.ok) {
+			const reader = res.body
+				.pipeThrough(new TextDecoderStream())
+				.pipeThrough(splitStream('\n'))
+				.getReader();
+
+			while (true) {
+				const { value, done } = await reader.read();
+				if (done || stopResponseFlag) {
+					currentRequestId = null;
+					break;
+				}
+
+				try {
+					let lines = value.split('\n');
+
+					for (const line of lines) {
+						if (line !== '') {
+							if (line === 'data: [DONE]') {
+								// responseMessage.done = true;
+								console.log('done');
+							} else {
+								let data = JSON.parse(line.replace(/^data: /, ''));
+								console.log(data);
+
+								text += data.choices[0].delta.content ?? '';
+							}
+						}
+					}
+				} catch (error) {
+					console.log(error);
+				}
+
+				scrollToBottom();
+			}
+		}
+	};
+
+	const chatCompletionHandler = async () => {
+		const model = $models.find((model) => model.id === selectedModelId);
+
+		const res = await generateOpenAIChatCompletion(
+			localStorage.token,
+			{
+				model: model.id,
+				stream: true,
+				messages: [
+					system
+						? {
+								role: 'system',
+								content: system
+						  }
+						: undefined,
+					...messages
+				].filter((message) => message)
+			},
+			model.external
+				? model.source === 'litellm'
+					? `${LITELLM_API_BASE_URL}/v1`
+					: `${OPENAI_API_BASE_URL}`
+				: `${OLLAMA_API_BASE_URL}/v1`
+		);
+
+		// const [res, controller] = await generateChatCompletion(localStorage.token, {
+		// 	model: selectedModelId,
+		// 	messages: [
+		// 		{
+		// 			role: 'assistant',
+		// 			content: text
+		// 		}
+		// 	]
+		// });
+
+		let responseMessage;
+		if (messages.at(-1)?.role === 'assistant') {
+			responseMessage = messages.at(-1);
+		} else {
+			responseMessage = {
+				role: 'assistant',
+				content: ''
+			};
+			messages.push(responseMessage);
+			messages = messages;
+		}
+
+		await tick();
+		const textareaElement = document.getElementById(`assistant-${messages.length - 1}-textarea`);
+
+		if (res && res.ok) {
+			const reader = res.body
+				.pipeThrough(new TextDecoderStream())
+				.pipeThrough(splitStream('\n'))
+				.getReader();
+
+			while (true) {
+				const { value, done } = await reader.read();
+				if (done || stopResponseFlag) {
+					break;
+				}
+
+				try {
+					let lines = value.split('\n');
+
+					for (const line of lines) {
+						if (line !== '') {
+							console.log(line);
+							if (line === 'data: [DONE]') {
+								// responseMessage.done = true;
+								messages = messages;
+							} else {
+								let data = JSON.parse(line.replace(/^data: /, ''));
+								console.log(data);
+
+								if (responseMessage.content == '' && data.choices[0].delta.content == '\n') {
+									continue;
+								} else {
+									textareaElement.style.height = textareaElement.scrollHeight + 'px';
+
+									responseMessage.content += data.choices[0].delta.content ?? '';
+									messages = messages;
+
+									textareaElement.style.height = textareaElement.scrollHeight + 'px';
+
+									await tick();
+								}
+							}
+						}
+					}
+				} catch (error) {
+					console.log(error);
+				}
+
+				scrollToBottom();
+			}
+
+			// while (true) {
+			// 	const { value, done } = await reader.read();
+			// 	if (done || stopResponseFlag) {
+			// 		if (stopResponseFlag) {
+			// 			await cancelChatCompletion(localStorage.token, currentRequestId);
+			// 		}
+
+			// 		currentRequestId = null;
+			// 		break;
+			// 	}
+
+			// 	try {
+			// 		let lines = value.split('\n');
+
+			// 		for (const line of lines) {
+			// 			if (line !== '') {
+			// 				console.log(line);
+			// 				let data = JSON.parse(line);
+
+			// 				if ('detail' in data) {
+			// 					throw data;
+			// 				}
+
+			// 				if ('id' in data) {
+			// 					console.log(data);
+			// 					currentRequestId = data.id;
+			// 				} else {
+			// 					if (data.done == false) {
+			// 						text += data.message.content;
+			// 					} else {
+			// 						console.log('done');
+			// 					}
+			// 				}
+			// 			}
+			// 		}
+			// 	} catch (error) {
+			// 		console.log(error);
+			// 	}
+
+			// 	scrollToBottom();
+			// }
+		}
+	};
+
+	const submitHandler = async () => {
+		if (selectedModelId) {
+			loading = true;
+
+			if (mode === 'complete') {
+				await textCompletionHandler();
+			} else if (mode === 'chat') {
+				await chatCompletionHandler();
+			}
+
+			loading = false;
+			stopResponseFlag = false;
+			currentRequestId = null;
+		}
+	};
+
+	onMount(async () => {
+		if ($user?.role !== 'admin') {
+			await goto('/');
+		}
+
+		if ($settings?.models) {
+			selectedModelId = $settings?.models[0];
+		} else if ($config?.default_models) {
+			selectedModelId = $config?.default_models.split(',')[0];
+		} else {
+			selectedModelId = '';
+		}
+		loaded = true;
+	});
+</script>
+
+<svelte:head>
+	<title>
+		{`Playground | ${$WEBUI_NAME}`}
+	</title>
+</svelte:head>
+
+<div class="min-h-screen max-h-[100dvh] w-full flex justify-center dark:text-white">
+	<div class=" flex flex-col justify-between w-full overflow-y-auto h-[100dvh]">
+		<div class="max-w-2xl mx-auto w-full px-3 p-3 md:px-0 h-full">
+			<div class=" flex flex-col h-full">
+				<div class="flex flex-col justify-between mb-2.5 gap-1">
+					<div class="flex justify-between items-center gap-2">
+						<div class=" text-2xl font-semibold self-center flex">
+							Playground <span class=" text-xs text-gray-500 self-center ml-1">(Beta)</span>
+						</div>
+
+						<div>
+							<button
+								class=" flex items-center gap-0.5 text-xs px-2.5 py-0.5 rounded-lg {mode ===
+									'chat' && 'text-sky-600 dark:text-sky-200 bg-sky-200/30'} {mode === 'complete' &&
+									'text-green-600 dark:text-green-200 bg-green-200/30'} "
+								on:click={() => {
+									if (mode === 'complete') {
+										mode = 'chat';
+									} else {
+										mode = 'complete';
+									}
+								}}
+							>
+								{#if mode === 'complete'}
+									Text Completion
+								{:else if mode === 'chat'}
+									Chat
+								{/if}
+
+								<div>
+									<svg
+										xmlns="http://www.w3.org/2000/svg"
+										viewBox="0 0 16 16"
+										fill="currentColor"
+										class="w-3 h-3"
+									>
+										<path
+											fill-rule="evenodd"
+											d="M5.22 10.22a.75.75 0 0 1 1.06 0L8 11.94l1.72-1.72a.75.75 0 1 1 1.06 1.06l-2.25 2.25a.75.75 0 0 1-1.06 0l-2.25-2.25a.75.75 0 0 1 0-1.06ZM10.78 5.78a.75.75 0 0 1-1.06 0L8 4.06 6.28 5.78a.75.75 0 0 1-1.06-1.06l2.25-2.25a.75.75 0 0 1 1.06 0l2.25 2.25a.75.75 0 0 1 0 1.06Z"
+											clip-rule="evenodd"
+										/>
+									</svg>
+								</div>
+							</button>
+						</div>
+					</div>
+
+					<div class="  flex gap-1 px-1">
+						<select
+							id="models"
+							class="outline-none bg-transparent text-sm font-medium rounded-lg w-full placeholder-gray-400"
+							bind:value={selectedModelId}
+						>
+							<option class=" text-gray-800" value="" selected disabled>Select a model</option>
+
+							{#each $models as model}
+								{#if model.name === 'hr'}
+									<hr />
+								{:else}
+									<option value={model.id} class="text-gray-800 text-lg"
+										>{model.name +
+											`${model.size ? ` (${(model.size / 1024 ** 3).toFixed(1)}GB)` : ''}`}</option
+									>
+								{/if}
+							{/each}
+						</select>
+
+						<!-- <button
+							class=" self-center dark:hover:text-gray-300"
+							id="open-settings-button"
+							on:click={async () => {}}
+						>
+							<svg
+								xmlns="http://www.w3.org/2000/svg"
+								fill="none"
+								viewBox="0 0 24 24"
+								stroke-width="1.5"
+								stroke="currentColor"
+								class="w-4 h-4"
+							>
+								<path
+									stroke-linecap="round"
+									stroke-linejoin="round"
+									d="M10.343 3.94c.09-.542.56-.94 1.11-.94h1.093c.55 0 1.02.398 1.11.94l.149.894c.07.424.384.764.78.93.398.164.855.142 1.205-.108l.737-.527a1.125 1.125 0 011.45.12l.773.774c.39.389.44 1.002.12 1.45l-.527.737c-.25.35-.272.806-.107 1.204.165.397.505.71.93.78l.893.15c.543.09.94.56.94 1.109v1.094c0 .55-.397 1.02-.94 1.11l-.893.149c-.425.07-.765.383-.93.78-.165.398-.143.854.107 1.204l.527.738c.32.447.269 1.06-.12 1.45l-.774.773a1.125 1.125 0 01-1.449.12l-.738-.527c-.35-.25-.806-.272-1.203-.107-.397.165-.71.505-.781.929l-.149.894c-.09.542-.56.94-1.11.94h-1.094c-.55 0-1.019-.398-1.11-.94l-.148-.894c-.071-.424-.384-.764-.781-.93-.398-.164-.854-.142-1.204.108l-.738.527c-.447.32-1.06.269-1.45-.12l-.773-.774a1.125 1.125 0 01-.12-1.45l.527-.737c.25-.35.273-.806.108-1.204-.165-.397-.505-.71-.93-.78l-.894-.15c-.542-.09-.94-.56-.94-1.109v-1.094c0-.55.398-1.02.94-1.11l.894-.149c.424-.07.765-.383.93-.78.165-.398.143-.854-.107-1.204l-.527-.738a1.125 1.125 0 01.12-1.45l.773-.773a1.125 1.125 0 011.45-.12l.737.527c.35.25.807.272 1.204.107.397-.165.71-.505.78-.929l.15-.894z"
+								/>
+								<path
+									stroke-linecap="round"
+									stroke-linejoin="round"
+									d="M15 12a3 3 0 11-6 0 3 3 0 016 0z"
+								/>
+							</svg>
+						</button> -->
+					</div>
+				</div>
+
+				{#if mode === 'chat'}
+					<div class="p-1">
+						<div class="p-3 outline outline-1 outline-gray-200 dark:outline-gray-800 rounded-lg">
+							<div class=" text-sm font-medium">System</div>
+							<textarea
+								id="system-textarea"
+								class="w-full h-full bg-transparent resize-none outline-none text-sm"
+								bind:value={system}
+								placeholder="You're a helpful assistant."
+								rows="4"
+							/>
+						</div>
+					</div>
+				{/if}
+
+				<div
+					class=" pb-2.5 flex flex-col justify-between w-full flex-auto overflow-auto h-0"
+					id="messages-container"
+				>
+					<div class=" h-full w-full flex flex-col">
+						<div class="flex-1 p-1">
+							{#if mode === 'complete'}
+								<textarea
+									id="text-completion-textarea"
+									class="w-full h-full p-3 bg-transparent outline outline-1 outline-gray-200 dark:outline-gray-800 resize-none rounded-lg text-sm"
+									bind:value={text}
+									placeholder="You're a helpful assistant."
+								/>
+							{:else if mode === 'chat'}
+								<ChatCompletion bind:messages />
+							{/if}
+						</div>
+					</div>
+				</div>
+
+				<div class="pb-2">
+					{#if !loading}
+						<button
+							class="px-3 py-1.5 text-sm font-medium bg-emerald-600 hover:bg-emerald-700 text-gray-50 transition rounded-lg"
+							on:click={() => {
+								submitHandler();
+							}}
+						>
+							Submit
+						</button>
+					{:else}
+						<button
+							class="px-3 py-1.5 text-sm font-medium bg-gray-100 hover:bg-gray-200 text-gray-900 transition rounded-lg"
+							on:click={() => {
+								stopResponse();
+							}}
+						>
+							Cancel
+						</button>
+					{/if}
+				</div>
+			</div>
+		</div>
+	</div>
+</div>
+
+<style>
+	.scrollbar-hidden::-webkit-scrollbar {
+		display: none; /* for Chrome, Safari and Opera */
+	}
+
+	.scrollbar-hidden {
+		-ms-overflow-style: none; /* IE and Edge */
+		scrollbar-width: none; /* Firefox */
+	}
+</style>