Merge branch 'dev' into feat/system-wide-theme

CHANGELOG.md

@@ -5,6 +5,42 @@ 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).
 
+## [0.1.113] - 2024-03-XX
+
+### Added
+
+- 🌍 **Localization**: You can now change the UI language in Settings -> General. We support Ukrainian, German, Farsi (Persian), Traditional and Simplified Chinese and French translations. You can help us to translate the UI into your language! More info in our [CONTRIBUTION.md](https://github.com/open-webui/open-webui/blob/main/docs/CONTRIBUTING.md#-translations-and-internationalization).
+
+### Fixed
+
+- 🌑 **Dark background on select fields**: Added dark background to select fields, as this caused bad readability on some browsers/devices.
+
+## [0.1.112] - 2024-03-15
+
+### Fixed
+
+- 🗨️ Resolved chat malfunction after image generation.
+- 🎨 Fixed various RAG issues.
+- 🧪 Rectified experimental broken GGUF upload logic.
+
+## [0.1.111] - 2024-03-10
+
+### Added
+
+- 🛡️ **Model Whitelisting**: Admins now have the ability to whitelist models for users with the 'user' role.
+- 🔄 **Update All Models**: Added a convenient button to update all models at once.
+- 📄 **Toggle PDF OCR**: Users can now toggle PDF OCR option for improved parsing performance.
+- 🎨 **DALL-E Integration**: Introduced DALL-E integration for image generation alongside automatic1111.
+- 🛠️ **RAG API Refactoring**: Refactored RAG logic and exposed its API, with additional documentation to follow.
+
+### Fixed
+
+- 🔒 **Max Token Settings**: Added max token settings for anthropic/claude-3-sonnet-20240229 (Issue #1094).
+- 🔧 **Misalignment Issue**: Corrected misalignment of Edit and Delete Icons when Chat Title is Empty (Issue #1104).
+- 🔄 **Context Loss Fix**: Resolved RAG losing context on model response regeneration with Groq models via API key (Issue #1105).
+- 📁 **File Handling Bug**: Addressed File Not Found Notification when Dropping a Conversation Element (Issue #1098).
+- 🖱️ **Dragged File Styling**: Fixed dragged file layover styling issue.
+
 ## [0.1.110] - 2024-03-06
 
 ### Added

README.md

@@ -11,7 +11,7 @@
 [![Discord](https://img.shields.io/badge/Discord-Open_WebUI-blue?logo=discord&logoColor=white)](https://discord.gg/5rJgQTnV4s)
 [![](https://img.shields.io/static/v1?label=Sponsor&message=%E2%9D%A4&logo=GitHub&color=%23fe8e86)](https://github.com/sponsors/tjbck)
 
-User-friendly WebUI for LLMs, supported LLM runners include Ollama and OpenAI-compatible APIs. For more information, be sure to check out our [Open WebUI Documentation](https://docs.openwebui.com/).
+Open WebUI is an extensible, feature-rich, and user-friendly self-hosted WebUI designed to operate entirely offline. It supports various LLM runners, including Ollama and OpenAI-compatible APIs. For more information, be sure to check out our [Open WebUI Documentation](https://docs.openwebui.com/).
 
 ![Open WebUI Demo](./demo.gif)
 

backend/apps/images/main.py

@@ -293,6 +293,7 @@ def generate_image(
                 "size": form_data.size if form_data.size else app.state.IMAGE_SIZE,
                 "response_format": "b64_json",
             }
+
             r = requests.post(
                 url=f"https://api.openai.com/v1/images/generations",
                 json=data,
@@ -300,7 +301,6 @@ def generate_image(
             )
 
             r.raise_for_status()
-
             res = r.json()
 
             images = []
@@ -356,7 +356,10 @@ def generate_image(
             return images
 
     except Exception as e:
-        print(e)
-        if r:
-            print(r.json())
-        raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
+        error = e
+
+        if r != None:
+            data = r.json()
+            if "error" in data:
+                error = data["error"]["message"]
+        raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(error))

backend/apps/ollama/main.py

@@ -15,7 +15,7 @@ import asyncio
 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_BASE_URLS
+from config import OLLAMA_BASE_URLS, MODEL_FILTER_ENABLED, MODEL_FILTER_LIST
 
 from typing import Optional, List, Union
 
@@ -29,6 +29,10 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+
+app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
+app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+
 app.state.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS
 app.state.MODELS = {}
 
@@ -119,6 +123,7 @@ async def get_all_models():
             map(lambda response: response["models"], responses)
         )
     }
+
     app.state.MODELS = {model["model"]: model for model in models["models"]}
 
     return models
@@ -129,9 +134,19 @@ async def get_all_models():
 async def get_ollama_tags(
     url_idx: Optional[int] = None, user=Depends(get_current_user)
 ):
-
     if url_idx == None:
-        return await get_all_models()
+        models = await get_all_models()
+
+        if app.state.MODEL_FILTER_ENABLED:
+            if user.role == "user":
+                models["models"] = list(
+                    filter(
+                        lambda model: model["name"] in app.state.MODEL_FILTER_LIST,
+                        models["models"],
+                    )
+                )
+                return models
+        return models
     else:
         url = app.state.OLLAMA_BASE_URLS[url_idx]
         try:
@@ -167,11 +182,17 @@ async def get_ollama_versions(url_idx: Optional[int] = None):
         responses = await asyncio.gather(*tasks)
         responses = list(filter(lambda x: x is not None, responses))
 
+        if len(responses) > 0:
             lowest_version = min(
                 responses, key=lambda x: tuple(map(int, x["version"].split(".")))
             )
 
             return {"version": lowest_version["version"]}
+        else:
+            raise HTTPException(
+                status_code=500,
+                detail=ERROR_MESSAGES.OLLAMA_NOT_FOUND,
+            )
     else:
         url = app.state.OLLAMA_BASE_URLS[url_idx]
         try:

backend/apps/openai/main.py

@@ -18,7 +18,13 @@ from utils.utils import (
     get_verified_user,
     get_admin_user,
 )
-from config import OPENAI_API_BASE_URLS, OPENAI_API_KEYS, CACHE_DIR
+from config import (
+    OPENAI_API_BASE_URLS,
+    OPENAI_API_KEYS,
+    CACHE_DIR,
+    MODEL_FILTER_ENABLED,
+    MODEL_FILTER_LIST,
+)
 from typing import List, Optional
 
 
@@ -34,6 +40,9 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
+app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+
 app.state.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS
 app.state.OPENAI_API_KEYS = OPENAI_API_KEYS
 
@@ -102,6 +111,7 @@ async def speech(request: Request, user=Depends(get_verified_user)):
         headers["Authorization"] = f"Bearer {app.state.OPENAI_API_KEYS[idx]}"
         headers["Content-Type"] = "application/json"
 
+        r = None
         try:
             r = requests.post(
                 url=f"{app.state.OPENAI_API_BASE_URLS[idx]}/audio/speech",
@@ -134,7 +144,9 @@ async def speech(request: Request, user=Depends(get_verified_user)):
                 except:
                     error_detail = f"External: {e}"
 
-            raise HTTPException(status_code=r.status_code, detail=error_detail)
+            raise HTTPException(
+                status_code=r.status_code if r else 500, detail=error_detail
+            )
 
     except ValueError:
         raise HTTPException(status_code=401, detail=ERROR_MESSAGES.OPENAI_NOT_FOUND)
@@ -170,12 +182,18 @@ def merge_models_lists(model_lists):
 
 async def get_all_models():
     print("get_all_models")
+
+    if len(app.state.OPENAI_API_KEYS) == 1 and app.state.OPENAI_API_KEYS[0] == "":
+        models = {"data": []}
+    else:
         tasks = [
             fetch_url(f"{url}/models", app.state.OPENAI_API_KEYS[idx])
             for idx, url in enumerate(app.state.OPENAI_API_BASE_URLS)
         ]
         responses = await asyncio.gather(*tasks)
-    responses = list(filter(lambda x: x is not None and "error" not in x, responses))
+        responses = list(
+            filter(lambda x: x is not None and "error" not in x, responses)
+        )
         models = {
             "data": merge_models_lists(
                 list(map(lambda response: response["data"], responses))
@@ -186,14 +204,26 @@ async def get_all_models():
         return models
 
 
-# , user=Depends(get_current_user)
 @app.get("/models")
 @app.get("/models/{url_idx}")
-async def get_models(url_idx: Optional[int] = None):
+async def get_models(url_idx: Optional[int] = None, user=Depends(get_current_user)):
     if url_idx == None:
-        return await get_all_models()
+        models = await get_all_models()
+        if app.state.MODEL_FILTER_ENABLED:
+            if user.role == "user":
+                models["data"] = list(
+                    filter(
+                        lambda model: model["id"] in app.state.MODEL_FILTER_LIST,
+                        models["data"],
+                    )
+                )
+                return models
+        return models
     else:
         url = app.state.OPENAI_API_BASE_URLS[url_idx]
+
+        r = None
+
         try:
             r = requests.request(method="GET", url=f"{url}/models")
             r.raise_for_status()
@@ -266,6 +296,8 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
     headers["Authorization"] = f"Bearer {key}"
     headers["Content-Type"] = "application/json"
 
+    r = None
+
     try:
         r = requests.request(
             method=request.method,
@@ -298,4 +330,6 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
             except:
                 error_detail = f"External: {e}"
 
-        raise HTTPException(status_code=r.status_code, detail=error_detail)
+        raise HTTPException(
+            status_code=r.status_code if r else 500, detail=error_detail
+        )

backend/apps/rag/main.py

@@ -77,6 +77,7 @@ from constants import ERROR_MESSAGES
 
 app = FastAPI()
 
+app.state.PDF_EXTRACT_IMAGES = False
 app.state.CHUNK_SIZE = CHUNK_SIZE
 app.state.CHUNK_OVERLAP = CHUNK_OVERLAP
 app.state.RAG_TEMPLATE = RAG_TEMPLATE
@@ -184,12 +185,15 @@ async def update_embedding_model(
     }
 
 
-@app.get("/chunk")
-async def get_chunk_params(user=Depends(get_admin_user)):
+@app.get("/config")
+async def get_rag_config(user=Depends(get_admin_user)):
     return {
         "status": True,
+        "pdf_extract_images": app.state.PDF_EXTRACT_IMAGES,
+        "chunk": {
             "chunk_size": app.state.CHUNK_SIZE,
             "chunk_overlap": app.state.CHUNK_OVERLAP,
+        },
     }
 
 
@@ -198,17 +202,24 @@ class ChunkParamUpdateForm(BaseModel):
     chunk_overlap: int
 
 
-@app.post("/chunk/update")
-async def update_chunk_params(
-    form_data: ChunkParamUpdateForm, user=Depends(get_admin_user)
-):
-    app.state.CHUNK_SIZE = form_data.chunk_size
-    app.state.CHUNK_OVERLAP = form_data.chunk_overlap
+class ConfigUpdateForm(BaseModel):
+    pdf_extract_images: bool
+    chunk: ChunkParamUpdateForm
+
+
+@app.post("/config/update")
+async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_user)):
+    app.state.PDF_EXTRACT_IMAGES = form_data.pdf_extract_images
+    app.state.CHUNK_SIZE = form_data.chunk.chunk_size
+    app.state.CHUNK_OVERLAP = form_data.chunk.chunk_overlap
 
     return {
         "status": True,
+        "pdf_extract_images": app.state.PDF_EXTRACT_IMAGES,
+        "chunk": {
             "chunk_size": app.state.CHUNK_SIZE,
             "chunk_overlap": app.state.CHUNK_OVERLAP,
+        },
     }
 
 
@@ -364,7 +375,7 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
     ]
 
     if file_ext == "pdf":
-        loader = PyPDFLoader(file_path, extract_images=True)
+        loader = PyPDFLoader(file_path, extract_images=app.state.PDF_EXTRACT_IMAGES)
     elif file_ext == "csv":
         loader = CSVLoader(file_path)
     elif file_ext == "rst":
@@ -389,9 +400,9 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
     elif file_ext in known_source_ext or (
         file_content_type and file_content_type.find("text/") >= 0
     ):
-        loader = TextLoader(file_path)
+        loader = TextLoader(file_path, autodetect_encoding=True)
     else:
-        loader = TextLoader(file_path)
+        loader = TextLoader(file_path, autodetect_encoding=True)
         known_type = False
 
     return loader, known_type

backend/apps/rag/utils.py

@@ -91,7 +91,92 @@ def query_collection(
 
 
 def rag_template(template: str, context: str, query: str):
-    template = re.sub(r"\[context\]", context, template)
-    template = re.sub(r"\[query\]", query, template)
-
+    template = template.replace("[context]", context)
+    template = template.replace("[query]", query)
     return template
+
+
+def rag_messages(docs, messages, template, k, embedding_function):
+    print(docs)
+
+    last_user_message_idx = None
+    for i in range(len(messages) - 1, -1, -1):
+        if messages[i]["role"] == "user":
+            last_user_message_idx = i
+            break
+
+    user_message = messages[last_user_message_idx]
+
+    if isinstance(user_message["content"], list):
+        # Handle list content input
+        content_type = "list"
+        query = ""
+        for content_item in user_message["content"]:
+            if content_item["type"] == "text":
+                query = content_item["text"]
+                break
+    elif isinstance(user_message["content"], str):
+        # Handle text content input
+        content_type = "text"
+        query = user_message["content"]
+    else:
+        # Fallback in case the input does not match expected types
+        content_type = None
+        query = ""
+
+    relevant_contexts = []
+
+    for doc in docs:
+        context = None
+
+        try:
+            if doc["type"] == "collection":
+                context = query_collection(
+                    collection_names=doc["collection_names"],
+                    query=query,
+                    k=k,
+                    embedding_function=embedding_function,
+                )
+            else:
+                context = query_doc(
+                    collection_name=doc["collection_name"],
+                    query=query,
+                    k=k,
+                    embedding_function=embedding_function,
+                )
+        except Exception as e:
+            print(e)
+            context = None
+
+        relevant_contexts.append(context)
+
+    context_string = ""
+    for context in relevant_contexts:
+        if context:
+            context_string += " ".join(context["documents"][0]) + "\n"
+
+    ra_content = rag_template(
+        template=template,
+        context=context_string,
+        query=query,
+    )
+
+    if content_type == "list":
+        new_content = []
+        for content_item in user_message["content"]:
+            if content_item["type"] == "text":
+                # Update the text item's content with ra_content
+                new_content.append({"type": "text", "text": ra_content})
+            else:
+                # Keep other types of content as they are
+                new_content.append(content_item)
+        new_user_message = {**user_message, "content": new_content}
+    else:
+        new_user_message = {
+            **user_message,
+            "content": ra_content,
+        }
+
+    messages[last_user_message_idx] = new_user_message
+
+    return messages

backend/apps/web/routers/utils.py

@@ -75,7 +75,7 @@ async def download_file_stream(url, file_path, file_name, chunk_size=1024 * 1024
                     hashed = calculate_sha256(file)
                     file.seek(0)
 
-                    url = f"{OLLAMA_BASE_URLS[0]}/blobs/sha256:{hashed}"
+                    url = f"{OLLAMA_BASE_URLS[0]}/api/blobs/sha256:{hashed}"
                     response = requests.post(url, data=file)
 
                     if response.ok:

backend/config.py

@@ -209,10 +209,6 @@ OLLAMA_API_BASE_URL = os.environ.get(
 
 OLLAMA_BASE_URL = os.environ.get("OLLAMA_BASE_URL", "")
 
-if ENV == "prod":
-    if OLLAMA_BASE_URL == "/ollama":
-        OLLAMA_BASE_URL = "http://host.docker.internal:11434"
-
 
 if OLLAMA_BASE_URL == "" and OLLAMA_API_BASE_URL != "":
     OLLAMA_BASE_URL = (
@@ -221,6 +217,11 @@ if OLLAMA_BASE_URL == "" and OLLAMA_API_BASE_URL != "":
         else OLLAMA_API_BASE_URL
     )
 
+if ENV == "prod":
+    if OLLAMA_BASE_URL == "/ollama":
+        OLLAMA_BASE_URL = "http://host.docker.internal:11434"
+
+
 OLLAMA_BASE_URLS = os.environ.get("OLLAMA_BASE_URLS", "")
 OLLAMA_BASE_URLS = OLLAMA_BASE_URLS if OLLAMA_BASE_URLS != "" else OLLAMA_BASE_URL
 
@@ -234,8 +235,6 @@ OLLAMA_BASE_URLS = [url.strip() for url in OLLAMA_BASE_URLS.split(";")]
 OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "")
 OPENAI_API_BASE_URL = os.environ.get("OPENAI_API_BASE_URL", "")
 
-if OPENAI_API_KEY == "":
-    OPENAI_API_KEY = "none"
 
 if OPENAI_API_BASE_URL == "":
     OPENAI_API_BASE_URL = "https://api.openai.com/v1"
@@ -292,6 +291,11 @@ DEFAULT_USER_ROLE = os.getenv("DEFAULT_USER_ROLE", "pending")
 USER_PERMISSIONS = {"chat": {"deletion": True}}
 
 
+MODEL_FILTER_ENABLED = os.environ.get("MODEL_FILTER_ENABLED", False)
+MODEL_FILTER_LIST = os.environ.get("MODEL_FILTER_LIST", "")
+MODEL_FILTER_LIST = [model.strip() for model in MODEL_FILTER_LIST.split(";")]
+
+
 ####################################
 # WEBUI_VERSION
 ####################################

backend/constants.py

@@ -52,3 +52,4 @@ class ERROR_MESSAGES(str, Enum):
 
     MODEL_NOT_FOUND = lambda name="": f"Model '{name}' was not found"
     OPENAI_NOT_FOUND = lambda name="": f"OpenAI API was not found"
+    OLLAMA_NOT_FOUND = "WebUI could not connect to Ollama"

backend/data/config.json

 {
+    "version": "0.0.1",
     "ui": {
         "prompt_suggestions": [
             {

backend/main.py

@@ -23,10 +23,22 @@ from apps.images.main import app as images_app
 from apps.rag.main import app as rag_app
 from apps.web.main import app as webui_app
 
+from pydantic import BaseModel
+from typing import List
 
-from apps.rag.utils import query_doc, query_collection, rag_template
 
-from config import WEBUI_NAME, ENV, VERSION, CHANGELOG, FRONTEND_BUILD_DIR
+from utils.utils import get_admin_user
+from apps.rag.utils import rag_messages
+
+from config import (
+    WEBUI_NAME,
+    ENV,
+    VERSION,
+    CHANGELOG,
+    FRONTEND_BUILD_DIR,
+    MODEL_FILTER_ENABLED,
+    MODEL_FILTER_LIST,
+)
 from constants import ERROR_MESSAGES
 
 
@@ -43,20 +55,10 @@ class SPAStaticFiles(StaticFiles):
 
 app = FastAPI(docs_url="/docs" if ENV == "dev" else None, redoc_url=None)
 
-origins = ["*"]
-
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=origins,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
+app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
+app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
 
-@app.on_event("startup")
-async def on_startup():
-    await litellm_app_startup()
+origins = ["*"]
 
 
 class RAGMiddleware(BaseHTTPMiddleware):
@@ -76,96 +78,33 @@ class RAGMiddleware(BaseHTTPMiddleware):
             # Example: Add a new key-value pair or modify existing ones
             # data["modified"] = True  # Example modification
             if "docs" in data:
-                docs = data["docs"]
-                print(docs)
-
-                last_user_message_idx = None
-                for i in range(len(data["messages"]) - 1, -1, -1):
-                    if data["messages"][i]["role"] == "user":
-                        last_user_message_idx = i
-                        break
-
-                user_message = data["messages"][last_user_message_idx]
-
-                if isinstance(user_message["content"], list):
-                    # Handle list content input
-                    content_type = "list"
-                    query = ""
-                    for content_item in user_message["content"]:
-                        if content_item["type"] == "text":
-                            query = content_item["text"]
-                            break
-                elif isinstance(user_message["content"], str):
-                    # Handle text content input
-                    content_type = "text"
-                    query = user_message["content"]
-                else:
-                    # Fallback in case the input does not match expected types
-                    content_type = None
-                    query = ""
-
-                relevant_contexts = []
 
-                for doc in docs:
-                    context = None
-
-                    try:
-                        if doc["type"] == "collection":
-                            context = query_collection(
-                                collection_names=doc["collection_names"],
-                                query=query,
-                                k=rag_app.state.TOP_K,
-                                embedding_function=rag_app.state.sentence_transformer_ef,
+                data = {**data}
+                data["messages"] = rag_messages(
+                    data["docs"],
+                    data["messages"],
+                    rag_app.state.RAG_TEMPLATE,
+                    rag_app.state.TOP_K,
+                    rag_app.state.sentence_transformer_ef,
                 )
-                        else:
-                            context = query_doc(
-                                collection_name=doc["collection_name"],
-                                query=query,
-                                k=rag_app.state.TOP_K,
-                                embedding_function=rag_app.state.sentence_transformer_ef,
-                            )
-                    except Exception as e:
-                        print(e)
-                        context = None
-
-                    relevant_contexts.append(context)
-
-                context_string = ""
-                for context in relevant_contexts:
-                    if context:
-                        context_string += " ".join(context["documents"][0]) + "\n"
-
-                ra_content = rag_template(
-                    template=rag_app.state.RAG_TEMPLATE,
-                    context=context_string,
-                    query=query,
-                )
-
-                if content_type == "list":
-                    new_content = []
-                    for content_item in user_message["content"]:
-                        if content_item["type"] == "text":
-                            # Update the text item's content with ra_content
-                            new_content.append({"type": "text", "text": ra_content})
-                        else:
-                            # Keep other types of content as they are
-                            new_content.append(content_item)
-                    new_user_message = {**user_message, "content": new_content}
-                else:
-                    new_user_message = {
-                        **user_message,
-                        "content": ra_content,
-                    }
-
-                data["messages"][last_user_message_idx] = new_user_message
                 del data["docs"]
 
+                print(data["messages"])
+
             modified_body_bytes = json.dumps(data).encode("utf-8")
 
-            # Create a new request with the modified body
-            scope = request.scope
-            scope["body"] = modified_body_bytes
-            request = Request(scope, receive=lambda: self._receive(modified_body_bytes))
+            # Replace the request body with the modified one
+            request._body = modified_body_bytes
+
+            # Set custom header to ensure content-length matches new body length
+            request.headers.__dict__["_list"] = [
+                (b"content-length", str(len(modified_body_bytes)).encode("utf-8")),
+                *[
+                    (k, v)
+                    for k, v in request.headers.raw
+                    if k.lower() != b"content-length"
+                ],
+            ]
 
         response = await call_next(request)
         return response
@@ -177,6 +116,15 @@ class RAGMiddleware(BaseHTTPMiddleware):
 app.add_middleware(RAGMiddleware)
 
 
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=origins,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
 @app.middleware("http")
 async def check_url(request: Request, call_next):
     start_time = int(time.time())
@@ -187,6 +135,11 @@ async def check_url(request: Request, call_next):
     return response
 
 
+@app.on_event("startup")
+async def on_startup():
+    await litellm_app_startup()
+
+
 app.mount("/api/v1", webui_app)
 app.mount("/litellm/api", litellm_app)
 
@@ -211,6 +164,39 @@ async def get_app_config():
     }
 
 
+@app.get("/api/config/model/filter")
+async def get_model_filter_config(user=Depends(get_admin_user)):
+    return {
+        "enabled": app.state.MODEL_FILTER_ENABLED,
+        "models": app.state.MODEL_FILTER_LIST,
+    }
+
+
+class ModelFilterConfigForm(BaseModel):
+    enabled: bool
+    models: List[str]
+
+
+@app.post("/api/config/model/filter")
+async def get_model_filter_config(
+    form_data: ModelFilterConfigForm, user=Depends(get_admin_user)
+):
+
+    app.state.MODEL_FILTER_ENABLED = form_data.enabled
+    app.state.MODEL_FILTER_LIST = form_data.models
+
+    ollama_app.state.MODEL_FILTER_ENABLED = app.state.MODEL_FILTER_ENABLED
+    ollama_app.state.MODEL_FILTER_LIST = app.state.MODEL_FILTER_LIST
+
+    openai_app.state.MODEL_FILTER_ENABLED = app.state.MODEL_FILTER_ENABLED
+    openai_app.state.MODEL_FILTER_LIST = app.state.MODEL_FILTER_LIST
+
+    return {
+        "enabled": app.state.MODEL_FILTER_ENABLED,
+        "models": app.state.MODEL_FILTER_LIST,
+    }
+
+
 @app.get("/api/version")
 async def get_app_config():
 

backend/requirements.txt

@@ -16,7 +16,8 @@ aiohttp
 peewee
 bcrypt
 
-litellm
+litellm==1.30.7
+argon2-cffi
 apscheduler
 google-generativeai
 

docs/CONTRIBUTING.md

@@ -50,6 +50,18 @@ We welcome pull requests. Before submitting one, please:
 
 Help us make Open WebUI more accessible by improving documentation, writing tutorials, or creating guides on setting up and optimizing the web UI.
 
+### 🌐 Translations and Internationalization
+
+Help us make Open WebUI available to a wider audience. In this section, we'll guide you through the process of adding new translations to the project.
+
+We use JSON files to store translations. You can find the existing translation files in the `src/lib/i18n/locales` directory. Each directory corresponds to a specific language, for example, `en-US` for English (US), `fr-FR` for French (France) and so on. You can refer to [ISO 639 Language Codes][http://www.lingoes.net/en/translator/langcode.htm] to find the appropriate code for a specific language.
+
+To add a new language:
+
+- Create a new directory in the `src/lib/i18n/locales` path with the appropriate language code as its name. For instance, if you're adding translations for Spanish (Spain), create a new directory named `es-ES`.
+- Copy the American English translation file(s) (from `en-US` directory in `src/lib/i18n/locale`) to this new directory and update the string values in JSON format according to your language. Make sure to preserve the structure of the JSON object.
+- Add the language code and its respective title to languages file at `src/lib/i18n/locales/languages.json`.
+
 ### 🤔 Questions & Feedback
 
 Got questions or feedback? Join our [Discord community](https://discord.gg/5rJgQTnV4s) or open an issue. We're here to help!

i18next-parser.config.ts

+// i18next-parser.config.ts
+import { getLanguages } from './src/lib/i18n/index.ts';
+
+const getLangCodes = async () => {
+	const languages = await getLanguages();
+	return languages.map((l) => l.code);
+};
+
+export default {
+	contextSeparator: '_',
+	createOldCatalogs: false,
+	defaultNamespace: 'translation',
+	defaultValue: '',
+	indentation: 2,
+	keepRemoved: false,
+	keySeparator: false,
+	lexers: {
+		svelte: ['JavascriptLexer'],
+		js: ['JavascriptLexer'],
+		ts: ['JavascriptLexer'],
+
+		default: ['JavascriptLexer']
+	},
+	lineEnding: 'auto',
+	locales: await getLangCodes(),
+	namespaceSeparator: false,
+	output: 'src/lib/i18n/locales/$LOCALE/$NAMESPACE.json',
+	pluralSeparator: '_',
+	input: 'src/**/*.{js,svelte}',
+	sort: true,
+	verbose: true,
+	failOnWarnings: false,
+	failOnUpdate: false,
+	customValueTemplate: null,
+	resetDefaultValueLocale: null,
+	i18nextOptions: null,
+	yamlOptions: null
+};

kubernetes/helm/templates/webui-service.yaml

@@ -4,6 +4,9 @@ metadata:
   name: {{ include "open-webui.name" . }}
   labels:
     {{- include "open-webui.labels" . | nindent 4 }}
+    {{- with .Values.webui.service.labels }}
+    {{- toYaml . | nindent 4 }}
+    {{- end }}
   {{- with .Values.webui.service.annotations }}
   annotations:
     {{- toYaml . | nindent 4 }}
@@ -11,14 +14,16 @@ metadata:
 spec:
   selector:
     {{- include "open-webui.selectorLabels" . | nindent 4 }}
-{{- with .Values.webui.service }}
-  type: {{ .type }}
+  type: {{ .Values.webui.service.type | default "ClusterIP" }}
   ports:
   - protocol: TCP
     name: http
-    port: {{ .port }}
+    port: {{ .Values.webui.service.port }}
     targetPort: http
-    {{- if .nodePort }}
-    nodePort: {{ .nodePort | int }}
+    {{- if .Values.webui.service.nodePort }}
+    nodePort: {{ .Values.webui.service.nodePort | int }}
+    {{- end }}
+  {{- if .Values.webui.service.loadBalancerClass }}
+  loadBalancerClass: {{ .Values.webui.service.loadBalancerClass | quote }}
   {{- end }}
-{{- end }}
+

kubernetes/helm/values.yaml

@@ -70,3 +70,5 @@ webui:
     port: 80
     containerPort: 8080
     nodePort: ""
+    labels: {}
+    loadBalancerClass: ""