open-webui/backend/apps/ollama/main.py

1334 lines
38 KiB
Python
Raw Normal View History

2024-03-22 07:45:00 +01:00
from fastapi import (
FastAPI,
Request,
Response,
HTTPException,
Depends,
status,
UploadFile,
File,
BackgroundTasks,
)
2024-01-04 22:06:31 +01:00
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import StreamingResponse
2024-01-06 02:16:35 +01:00
from fastapi.concurrency import run_in_threadpool
2023-11-15 01:28:51 +01:00
2024-03-05 10:41:22 +01:00
from pydantic import BaseModel, ConfigDict
2024-03-05 09:59:35 +01:00
2024-03-22 07:45:00 +01:00
import os
2024-03-22 08:55:59 +01:00
import copy
2024-03-05 09:59:35 +01:00
import random
2023-11-15 01:28:51 +01:00
import requests
import json
2024-01-18 04:19:44 +01:00
import uuid
2024-03-05 09:59:35 +01:00
import aiohttp
import asyncio
import logging
2024-03-22 07:45:00 +01:00
from urllib.parse import urlparse
from typing import Optional, List, Union
2023-11-15 01:28:51 +01:00
2023-11-19 01:47:12 +01:00
from apps.web.models.users import Users
from constants import ERROR_MESSAGES
from utils.utils import decode_token, get_current_user, get_admin_user
2023-11-15 01:28:51 +01:00
2024-03-22 07:45:00 +01:00
2024-03-31 10:13:39 +02:00
from config import (
SRC_LOG_LEVELS,
OLLAMA_BASE_URLS,
ENABLE_MODEL_FILTER,
2024-03-31 10:13:39 +02:00
MODEL_FILTER_LIST,
UPLOAD_DIR,
)
2024-03-25 00:04:03 +01:00
from utils.misc import calculate_sha256
2024-03-05 09:59:35 +01:00
log = logging.getLogger(__name__)
log.setLevel(SRC_LOG_LEVELS["OLLAMA"])
2024-03-05 09:59:35 +01:00
2024-01-04 22:06:31 +01:00
app = FastAPI()
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
2023-11-15 01:28:51 +01:00
2024-03-10 06:19:20 +01:00
app.state.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
2024-03-10 06:47:01 +01:00
app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
2024-03-10 06:19:20 +01:00
2024-03-06 20:44:00 +01:00
app.state.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS
2024-03-05 09:59:35 +01:00
app.state.MODELS = {}
2023-11-15 01:28:51 +01:00
2024-01-18 04:19:44 +01:00
REQUEST_POOL = []
2024-03-05 10:21:50 +01:00
# TODO: Implement a more intelligent load balancing mechanism for distributing requests among multiple backend instances.
# Current implementation uses a simple round-robin approach (random.choice). Consider incorporating algorithms like weighted round-robin,
# least connections, or least response time for better resource utilization and performance optimization.
2024-03-05 09:59:35 +01:00
@app.middleware("http")
async def check_url(request: Request, call_next):
if len(app.state.MODELS) == 0:
await get_all_models()
else:
pass
response = await call_next(request)
return response
2024-04-02 11:03:13 +02:00
@app.head("/")
@app.get("/")
async def get_status():
return {"status": True}
2024-03-05 09:59:35 +01:00
@app.get("/urls")
async def get_ollama_api_urls(user=Depends(get_admin_user)):
return {"OLLAMA_BASE_URLS": app.state.OLLAMA_BASE_URLS}
2024-01-04 22:06:31 +01:00
2023-11-15 01:28:51 +01:00
2024-01-04 22:06:31 +01:00
class UrlUpdateForm(BaseModel):
2024-03-05 09:59:35 +01:00
urls: List[str]
2024-01-04 22:06:31 +01:00
2024-03-05 09:59:35 +01:00
@app.post("/urls/update")
2024-02-17 08:30:38 +01:00
async def update_ollama_api_url(form_data: UrlUpdateForm, user=Depends(get_admin_user)):
2024-03-05 09:59:35 +01:00
app.state.OLLAMA_BASE_URLS = form_data.urls
log.info(f"app.state.OLLAMA_BASE_URLS: {app.state.OLLAMA_BASE_URLS}")
2024-03-05 09:59:35 +01:00
return {"OLLAMA_BASE_URLS": app.state.OLLAMA_BASE_URLS}
2024-01-05 10:25:34 +01:00
2024-01-18 04:19:44 +01:00
@app.get("/cancel/{request_id}")
async def cancel_ollama_request(request_id: str, user=Depends(get_current_user)):
if user:
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
return True
else:
raise HTTPException(status_code=401, detail=ERROR_MESSAGES.ACCESS_PROHIBITED)
2024-03-05 09:59:35 +01:00
async def fetch_url(url):
try:
async with aiohttp.ClientSession() as session:
async with session.get(url) as response:
return await response.json()
except Exception as e:
# Handle connection error here
log.error(f"Connection error: {e}")
2024-03-05 09:59:35 +01:00
return None
def merge_models_lists(model_lists):
merged_models = {}
for idx, model_list in enumerate(model_lists):
2024-03-18 09:11:48 +01:00
if model_list is not None:
for model in model_list:
digest = model["digest"]
if digest not in merged_models:
model["urls"] = [idx]
merged_models[digest] = model
else:
merged_models[digest]["urls"].append(idx)
2024-03-05 09:59:35 +01:00
return list(merged_models.values())
# user=Depends(get_current_user)
async def get_all_models():
log.info("get_all_models()")
2024-03-05 09:59:35 +01:00
tasks = [fetch_url(f"{url}/api/tags") for url in app.state.OLLAMA_BASE_URLS]
responses = await asyncio.gather(*tasks)
models = {
"models": merge_models_lists(
2024-03-18 09:11:48 +01:00
map(lambda response: response["models"] if response else None, responses)
2024-03-05 09:59:35 +01:00
)
}
2024-03-12 08:26:14 +01:00
2024-03-05 09:59:35 +01:00
app.state.MODELS = {model["model"]: model for model in models["models"]}
return models
@app.get("/api/tags")
@app.get("/api/tags/{url_idx}")
async def get_ollama_tags(
url_idx: Optional[int] = None, user=Depends(get_current_user)
):
if url_idx == None:
2024-03-10 06:19:20 +01:00
models = await get_all_models()
if app.state.ENABLE_MODEL_FILTER:
2024-03-10 06:19:20 +01:00
if user.role == "user":
models["models"] = list(
filter(
2024-03-10 06:47:01 +01:00
lambda model: model["name"] in app.state.MODEL_FILTER_LIST,
models["models"],
)
2024-03-10 06:19:20 +01:00
)
return models
return models
2024-03-05 09:59:35 +01:00
else:
url = app.state.OLLAMA_BASE_URLS[url_idx]
try:
r = requests.request(method="GET", url=f"{url}/api/tags")
r.raise_for_status()
return r.json()
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
@app.get("/api/version")
@app.get("/api/version/{url_idx}")
async def get_ollama_versions(url_idx: Optional[int] = None):
if url_idx == None:
# returns lowest version
tasks = [fetch_url(f"{url}/api/version") for url in app.state.OLLAMA_BASE_URLS]
responses = await asyncio.gather(*tasks)
responses = list(filter(lambda x: x is not None, responses))
2024-03-12 08:26:14 +01:00
if len(responses) > 0:
lowest_version = min(
2024-04-07 03:55:51 +02:00
responses,
key=lambda x: tuple(map(int, x["version"].split("-")[0].split("."))),
2024-03-12 08:26:14 +01:00
)
2024-03-05 09:59:35 +01:00
2024-03-12 08:26:14 +01:00
return {"version": lowest_version["version"]}
else:
raise HTTPException(
status_code=500,
detail=ERROR_MESSAGES.OLLAMA_NOT_FOUND,
)
2024-03-05 09:59:35 +01:00
else:
url = app.state.OLLAMA_BASE_URLS[url_idx]
try:
r = requests.request(method="GET", url=f"{url}/api/version")
r.raise_for_status()
return r.json()
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class ModelNameForm(BaseModel):
name: str
@app.post("/api/pull")
@app.post("/api/pull/{url_idx}")
async def pull_model(
form_data: ModelNameForm, url_idx: int = 0, user=Depends(get_admin_user)
):
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 11:12:55 +01:00
2024-03-05 09:59:35 +01:00
r = None
2024-03-05 11:12:55 +01:00
def get_request():
nonlocal url
2024-03-05 09:59:35 +01:00
nonlocal r
2024-03-23 21:12:54 +01:00
request_id = str(uuid.uuid4())
2024-03-05 09:59:35 +01:00
try:
2024-03-23 21:12:54 +01:00
REQUEST_POOL.append(request_id)
2024-03-05 09:59:35 +01:00
def stream_content():
2024-03-23 21:12:54 +01:00
try:
yield json.dumps({"id": request_id, "done": False}) + "\n"
for chunk in r.iter_content(chunk_size=8192):
if request_id in REQUEST_POOL:
yield chunk
else:
log.warning("User: canceled request")
2024-03-23 21:12:54 +01:00
break
finally:
if hasattr(r, "close"):
r.close()
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
2024-03-05 09:59:35 +01:00
r = requests.request(
method="POST",
url=f"{url}/api/pull",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
stream=True,
)
r.raise_for_status()
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
try:
2024-03-05 11:12:55 +01:00
return await run_in_threadpool(get_request)
2024-03-23 21:12:54 +01:00
2024-03-05 09:59:35 +01:00
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class PushModelForm(BaseModel):
name: str
insecure: Optional[bool] = None
stream: Optional[bool] = None
@app.delete("/api/push")
@app.delete("/api/push/{url_idx}")
async def push_model(
form_data: PushModelForm,
url_idx: Optional[int] = None,
user=Depends(get_admin_user),
):
if url_idx == None:
if form_data.name in app.state.MODELS:
url_idx = app.state.MODELS[form_data.name]["urls"][0]
else:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.debug(f"url: {url}")
2024-03-05 09:59:35 +01:00
r = None
def get_request():
nonlocal url
nonlocal r
try:
def stream_content():
for chunk in r.iter_content(chunk_size=8192):
yield chunk
r = requests.request(
method="POST",
url=f"{url}/api/push",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
)
r.raise_for_status()
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
try:
return await run_in_threadpool(get_request)
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class CreateModelForm(BaseModel):
name: str
modelfile: Optional[str] = None
stream: Optional[bool] = None
path: Optional[str] = None
@app.post("/api/create")
@app.post("/api/create/{url_idx}")
async def create_model(
form_data: CreateModelForm, url_idx: int = 0, user=Depends(get_admin_user)
):
log.debug(f"form_data: {form_data}")
2024-03-05 09:59:35 +01:00
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 11:19:25 +01:00
2024-03-05 09:59:35 +01:00
r = None
def get_request():
nonlocal url
nonlocal r
try:
def stream_content():
for chunk in r.iter_content(chunk_size=8192):
yield chunk
r = requests.request(
method="POST",
url=f"{url}/api/create",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
stream=True,
)
r.raise_for_status()
log.debug(f"r: {r}")
2024-03-05 09:59:35 +01:00
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
try:
return await run_in_threadpool(get_request)
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class CopyModelForm(BaseModel):
source: str
destination: str
@app.post("/api/copy")
@app.post("/api/copy/{url_idx}")
async def copy_model(
form_data: CopyModelForm,
url_idx: Optional[int] = None,
user=Depends(get_admin_user),
):
if url_idx == None:
if form_data.source in app.state.MODELS:
url_idx = app.state.MODELS[form_data.source]["urls"][0]
else:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.source),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
try:
r = requests.request(
method="POST",
url=f"{url}/api/copy",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
)
r.raise_for_status()
log.debug(f"r.text: {r.text}")
2024-03-05 09:59:35 +01:00
return True
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
@app.delete("/api/delete")
@app.delete("/api/delete/{url_idx}")
async def delete_model(
form_data: ModelNameForm,
url_idx: Optional[int] = None,
user=Depends(get_admin_user),
):
if url_idx == None:
if form_data.name in app.state.MODELS:
url_idx = app.state.MODELS[form_data.name]["urls"][0]
else:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
try:
r = requests.request(
method="DELETE",
url=f"{url}/api/delete",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
)
r.raise_for_status()
log.debug(f"r.text: {r.text}")
2024-03-05 09:59:35 +01:00
return True
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
@app.post("/api/show")
async def show_model_info(form_data: ModelNameForm, user=Depends(get_current_user)):
if form_data.name not in app.state.MODELS:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
2024-03-05 09:59:35 +01:00
)
url_idx = random.choice(app.state.MODELS[form_data.name]["urls"])
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
try:
r = requests.request(
method="POST",
url=f"{url}/api/show",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
)
r.raise_for_status()
return r.json()
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class GenerateEmbeddingsForm(BaseModel):
model: str
prompt: str
options: Optional[dict] = None
keep_alive: Optional[Union[int, str]] = None
@app.post("/api/embeddings")
@app.post("/api/embeddings/{url_idx}")
async def generate_embeddings(
form_data: GenerateEmbeddingsForm,
url_idx: Optional[int] = None,
user=Depends(get_current_user),
):
if url_idx == None:
2024-04-14 22:51:13 +02:00
model = form_data.model
if ":" not in model:
model = f"{model}:latest"
if model in app.state.MODELS:
2024-04-14 22:52:59 +02:00
url_idx = random.choice(app.state.MODELS[model]["urls"])
2024-03-05 09:59:35 +01:00
else:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
try:
r = requests.request(
method="POST",
url=f"{url}/api/embeddings",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
)
r.raise_for_status()
return r.json()
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
2024-04-14 23:55:00 +02:00
def generate_ollama_embeddings(
form_data: GenerateEmbeddingsForm,
url_idx: Optional[int] = None,
):
2024-04-15 00:47:45 +02:00
2024-04-15 01:48:15 +02:00
log.info(f"generate_ollama_embeddings {form_data}")
2024-04-15 00:47:45 +02:00
2024-04-14 23:55:00 +02:00
if url_idx == None:
model = form_data.model
if ":" not in model:
model = f"{model}:latest"
if model in app.state.MODELS:
url_idx = random.choice(app.state.MODELS[model]["urls"])
else:
raise HTTPException(
status_code=400,
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
try:
r = requests.request(
method="POST",
url=f"{url}/api/embeddings",
data=form_data.model_dump_json(exclude_none=True).encode(),
)
r.raise_for_status()
data = r.json()
2024-04-15 01:48:15 +02:00
log.info(f"generate_ollama_embeddings {data}")
2024-04-15 00:47:45 +02:00
2024-04-14 23:55:00 +02:00
if "embedding" in data:
return data["embedding"]
else:
raise "Something went wrong :/"
except Exception as e:
log.exception(e)
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise error_detail
2024-03-05 09:59:35 +01:00
class GenerateCompletionForm(BaseModel):
model: str
prompt: str
images: Optional[List[str]] = None
format: Optional[str] = None
options: Optional[dict] = None
system: Optional[str] = None
template: Optional[str] = None
context: Optional[str] = None
stream: Optional[bool] = True
raw: Optional[bool] = None
keep_alive: Optional[Union[int, str]] = None
@app.post("/api/generate")
@app.post("/api/generate/{url_idx}")
async def generate_completion(
form_data: GenerateCompletionForm,
url_idx: Optional[int] = None,
user=Depends(get_current_user),
):
if url_idx == None:
2024-04-14 22:51:13 +02:00
model = form_data.model
if ":" not in model:
model = f"{model}:latest"
if model in app.state.MODELS:
2024-04-14 22:52:59 +02:00
url_idx = random.choice(app.state.MODELS[model]["urls"])
2024-03-05 09:59:35 +01:00
else:
raise HTTPException(
status_code=400,
2024-03-31 22:40:57 +02:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
r = None
def get_request():
nonlocal form_data
nonlocal r
request_id = str(uuid.uuid4())
try:
REQUEST_POOL.append(request_id)
def stream_content():
try:
if form_data.stream:
yield json.dumps({"id": request_id, "done": False}) + "\n"
for chunk in r.iter_content(chunk_size=8192):
if request_id in REQUEST_POOL:
yield chunk
else:
log.warning("User: canceled request")
2024-03-05 09:59:35 +01:00
break
finally:
if hasattr(r, "close"):
r.close()
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
r = requests.request(
method="POST",
url=f"{url}/api/generate",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
stream=True,
)
r.raise_for_status()
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
try:
return await run_in_threadpool(get_request)
except Exception as e:
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
)
class ChatMessage(BaseModel):
role: str
content: str
images: Optional[List[str]] = None
class GenerateChatCompletionForm(BaseModel):
model: str
messages: List[ChatMessage]
format: Optional[str] = None
options: Optional[dict] = None
template: Optional[str] = None
2024-03-19 02:14:05 +01:00
stream: Optional[bool] = None
2024-03-05 09:59:35 +01:00
keep_alive: Optional[Union[int, str]] = None
@app.post("/api/chat")
@app.post("/api/chat/{url_idx}")
2024-03-05 10:41:22 +01:00
async def generate_chat_completion(
2024-03-05 09:59:35 +01:00
form_data: GenerateChatCompletionForm,
url_idx: Optional[int] = None,
user=Depends(get_current_user),
):
if url_idx == None:
2024-04-14 22:51:13 +02:00
model = form_data.model
if ":" not in model:
model = f"{model}:latest"
if model in app.state.MODELS:
2024-04-14 22:52:59 +02:00
url_idx = random.choice(app.state.MODELS[model]["urls"])
2024-03-05 09:59:35 +01:00
else:
raise HTTPException(
status_code=400,
2024-03-05 10:07:59 +01:00
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
2024-03-05 09:59:35 +01:00
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 09:59:35 +01:00
r = None
2024-03-31 10:13:39 +02:00
log.debug(
"form_data.model_dump_json(exclude_none=True).encode(): {0} ".format(
form_data.model_dump_json(exclude_none=True).encode()
)
)
2024-03-05 09:59:35 +01:00
def get_request():
nonlocal form_data
nonlocal r
request_id = str(uuid.uuid4())
try:
REQUEST_POOL.append(request_id)
def stream_content():
try:
if form_data.stream:
yield json.dumps({"id": request_id, "done": False}) + "\n"
for chunk in r.iter_content(chunk_size=8192):
if request_id in REQUEST_POOL:
yield chunk
else:
log.warning("User: canceled request")
2024-03-05 09:59:35 +01:00
break
finally:
if hasattr(r, "close"):
r.close()
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
r = requests.request(
method="POST",
url=f"{url}/api/chat",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 09:59:35 +01:00
stream=True,
)
r.raise_for_status()
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
log.exception(e)
2024-03-05 09:59:35 +01:00
raise e
try:
return await run_in_threadpool(get_request)
except Exception as e:
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
2024-03-05 10:41:22 +01:00
)
# TODO: we should update this part once Ollama supports other types
class OpenAIChatMessage(BaseModel):
role: str
content: str
model_config = ConfigDict(extra="allow")
class OpenAIChatCompletionForm(BaseModel):
model: str
messages: List[OpenAIChatMessage]
model_config = ConfigDict(extra="allow")
@app.post("/v1/chat/completions")
@app.post("/v1/chat/completions/{url_idx}")
async def generate_openai_chat_completion(
form_data: OpenAIChatCompletionForm,
url_idx: Optional[int] = None,
user=Depends(get_current_user),
):
if url_idx == None:
2024-04-14 22:51:13 +02:00
model = form_data.model
if ":" not in model:
model = f"{model}:latest"
if model in app.state.MODELS:
2024-04-14 22:52:59 +02:00
url_idx = random.choice(app.state.MODELS[model]["urls"])
2024-03-05 10:41:22 +01:00
else:
raise HTTPException(
status_code=400,
detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
)
url = app.state.OLLAMA_BASE_URLS[url_idx]
log.info(f"url: {url}")
2024-03-05 10:41:22 +01:00
r = None
def get_request():
nonlocal form_data
nonlocal r
request_id = str(uuid.uuid4())
try:
REQUEST_POOL.append(request_id)
def stream_content():
try:
if form_data.stream:
yield json.dumps(
{"request_id": request_id, "done": False}
) + "\n"
for chunk in r.iter_content(chunk_size=8192):
if request_id in REQUEST_POOL:
yield chunk
else:
log.warning("User: canceled request")
2024-03-05 10:41:22 +01:00
break
finally:
if hasattr(r, "close"):
r.close()
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
r = requests.request(
method="POST",
url=f"{url}/v1/chat/completions",
2024-03-07 03:37:40 +01:00
data=form_data.model_dump_json(exclude_none=True).encode(),
2024-03-05 10:41:22 +01:00
stream=True,
)
r.raise_for_status()
return StreamingResponse(
stream_content(),
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
try:
return await run_in_threadpool(get_request)
except Exception as e:
error_detail = "Open WebUI: Server Connection Error"
if r is not None:
try:
res = r.json()
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
raise HTTPException(
status_code=r.status_code if r else 500,
detail=error_detail,
2024-03-05 09:59:35 +01:00
)
2024-03-22 07:45:00 +01:00
class UrlForm(BaseModel):
url: str
class UploadBlobForm(BaseModel):
filename: str
def parse_huggingface_url(hf_url):
try:
# Parse the URL
parsed_url = urlparse(hf_url)
# Get the path and split it into components
path_components = parsed_url.path.split("/")
# Extract the desired output
user_repo = "/".join(path_components[1:3])
model_file = path_components[-1]
return model_file
except ValueError:
return None
async def download_file_stream(
ollama_url, file_url, file_path, file_name, chunk_size=1024 * 1024
):
done = False
if os.path.exists(file_path):
current_size = os.path.getsize(file_path)
else:
current_size = 0
headers = {"Range": f"bytes={current_size}-"} if current_size > 0 else {}
timeout = aiohttp.ClientTimeout(total=600) # Set the timeout
async with aiohttp.ClientSession(timeout=timeout) as session:
async with session.get(file_url, headers=headers) as response:
total_size = int(response.headers.get("content-length", 0)) + current_size
with open(file_path, "ab+") as file:
async for data in response.content.iter_chunked(chunk_size):
current_size += len(data)
file.write(data)
done = current_size == total_size
progress = round((current_size / total_size) * 100, 2)
2024-03-22 08:10:55 +01:00
2024-03-22 07:45:00 +01:00
yield f'data: {{"progress": {progress}, "completed": {current_size}, "total": {total_size}}}\n\n'
if done:
file.seek(0)
hashed = calculate_sha256(file)
file.seek(0)
url = f"{ollama_url}/api/blobs/sha256:{hashed}"
response = requests.post(url, data=file)
if response.ok:
res = {
"done": done,
"blob": f"sha256:{hashed}",
"name": file_name,
}
os.remove(file_path)
yield f"data: {json.dumps(res)}\n\n"
else:
raise "Ollama: Could not create blob, Please try again."
2024-03-22 08:10:55 +01:00
# def number_generator():
# for i in range(1, 101):
# yield f"data: {i}\n"
2024-03-22 07:45:00 +01:00
# url = "https://huggingface.co/TheBloke/stablelm-zephyr-3b-GGUF/resolve/main/stablelm-zephyr-3b.Q2_K.gguf"
@app.post("/models/download")
@app.post("/models/download/{url_idx}")
async def download_model(
form_data: UrlForm,
url_idx: Optional[int] = None,
):
2024-04-01 23:01:05 +02:00
allowed_hosts = ["https://huggingface.co/", "https://github.com/"]
if not any(form_data.url.startswith(host) for host in allowed_hosts):
raise HTTPException(
status_code=400,
detail="Invalid file_url. Only URLs from allowed hosts are permitted.",
)
2024-03-22 07:45:00 +01:00
if url_idx == None:
url_idx = 0
url = app.state.OLLAMA_BASE_URLS[url_idx]
file_name = parse_huggingface_url(form_data.url)
if file_name:
file_path = f"{UPLOAD_DIR}/{file_name}"
2024-04-01 23:01:05 +02:00
2024-03-22 07:45:00 +01:00
return StreamingResponse(
2024-03-22 08:10:55 +01:00
download_file_stream(url, form_data.url, file_path, file_name),
2024-03-22 07:45:00 +01:00
)
else:
return None
@app.post("/models/upload")
@app.post("/models/upload/{url_idx}")
def upload_model(file: UploadFile = File(...), url_idx: Optional[int] = None):
if url_idx == None:
url_idx = 0
ollama_url = app.state.OLLAMA_BASE_URLS[url_idx]
file_path = f"{UPLOAD_DIR}/{file.filename}"
# Save file in chunks
with open(file_path, "wb+") as f:
for chunk in file.file:
f.write(chunk)
def file_process_stream():
nonlocal ollama_url
total_size = os.path.getsize(file_path)
chunk_size = 1024 * 1024
try:
with open(file_path, "rb") as f:
total = 0
done = False
while not done:
chunk = f.read(chunk_size)
if not chunk:
done = True
continue
total += len(chunk)
progress = round((total / total_size) * 100, 2)
res = {
"progress": progress,
"total": total_size,
"completed": total,
}
yield f"data: {json.dumps(res)}\n\n"
if done:
f.seek(0)
hashed = calculate_sha256(f)
f.seek(0)
2024-03-22 08:58:38 +01:00
url = f"{ollama_url}/api/blobs/sha256:{hashed}"
2024-03-22 07:45:00 +01:00
response = requests.post(url, data=f)
if response.ok:
res = {
"done": done,
"blob": f"sha256:{hashed}",
"name": file.filename,
}
os.remove(file_path)
yield f"data: {json.dumps(res)}\n\n"
else:
raise Exception(
"Ollama: Could not create blob, Please try again."
)
except Exception as e:
res = {"error": str(e)}
yield f"data: {json.dumps(res)}\n\n"
return StreamingResponse(file_process_stream(), media_type="text/event-stream")
2024-03-22 08:55:59 +01:00
# async def upload_model(file: UploadFile = File(), url_idx: Optional[int] = None):
# if url_idx == None:
# url_idx = 0
# url = app.state.OLLAMA_BASE_URLS[url_idx]
# file_location = os.path.join(UPLOAD_DIR, file.filename)
# total_size = file.size
# async def file_upload_generator(file):
# print(file)
# try:
# async with aiofiles.open(file_location, "wb") as f:
# completed_size = 0
# while True:
# chunk = await file.read(1024*1024)
# if not chunk:
# break
# await f.write(chunk)
# completed_size += len(chunk)
# progress = (completed_size / total_size) * 100
# print(progress)
# yield f'data: {json.dumps({"status": "uploading", "percentage": progress, "total": total_size, "completed": completed_size, "done": False})}\n'
# except Exception as e:
# print(e)
# yield f"data: {json.dumps({'status': 'error', 'message': str(e)})}\n"
# finally:
# await file.close()
# print("done")
# yield f'data: {json.dumps({"status": "completed", "percentage": 100, "total": total_size, "completed": completed_size, "done": True})}\n'
# return StreamingResponse(
# file_upload_generator(copy.deepcopy(file)), media_type="text/event-stream"
# )
2024-01-04 22:06:31 +01:00
@app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
2024-03-05 10:21:50 +01:00
async def deprecated_proxy(path: str, request: Request, user=Depends(get_current_user)):
2024-03-05 09:59:35 +01:00
url = app.state.OLLAMA_BASE_URLS[0]
target_url = f"{url}/{path}"
2024-01-04 22:06:31 +01:00
body = await request.body()
headers = dict(request.headers)
if user.role in ["user", "admin"]:
if path in ["pull", "delete", "push", "copy", "create"]:
if user.role != "admin":
2024-01-05 10:25:34 +01:00
raise HTTPException(
2024-02-17 08:30:38 +01:00
status_code=status.HTTP_401_UNAUTHORIZED,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
2024-01-05 10:25:34 +01:00
)
2024-01-04 22:06:31 +01:00
else:
2024-02-17 08:30:38 +01:00
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
)
2023-12-15 02:05:46 +01:00
headers.pop("host", None)
headers.pop("authorization", None)
headers.pop("origin", None)
headers.pop("referer", None)
2023-12-26 22:40:03 +01:00
2024-01-06 02:16:35 +01:00
r = None
def get_request():
nonlocal r
2024-01-18 04:19:44 +01:00
request_id = str(uuid.uuid4())
2024-01-06 02:16:35 +01:00
try:
2024-01-18 04:19:44 +01:00
REQUEST_POOL.append(request_id)
def stream_content():
try:
2024-03-02 12:01:44 +01:00
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":
2024-01-18 04:19:44 +01:00
yield json.dumps({"id": request_id, "done": False}) + "\n"
for chunk in r.iter_content(chunk_size=8192):
if request_id in REQUEST_POOL:
yield chunk
else:
log.warning("User: canceled request")
2024-01-18 04:19:44 +01:00
break
finally:
if hasattr(r, "close"):
r.close()
2024-03-02 12:01:44 +01:00
if request_id in REQUEST_POOL:
REQUEST_POOL.remove(request_id)
2024-01-18 04:19:44 +01:00
2024-01-06 02:16:35 +01:00
r = requests.request(
method=request.method,
url=target_url,
data=body,
headers=headers,
stream=True,
)
r.raise_for_status()
2024-01-18 04:19:44 +01:00
# r.close()
2024-01-06 02:16:35 +01:00
return StreamingResponse(
2024-01-18 04:19:44 +01:00
stream_content(),
2024-01-06 02:16:35 +01:00
status_code=r.status_code,
headers=dict(r.headers),
)
except Exception as e:
raise e
2024-01-06 02:16:35 +01:00
try:
return await run_in_threadpool(get_request)
except Exception as e:
2024-02-17 08:30:38 +01:00
error_detail = "Open WebUI: Server Connection Error"
2024-01-06 02:16:35 +01:00
if r is not None:
2024-01-04 22:06:31 +01:00
try:
2024-01-06 02:16:35 +01:00
res = r.json()
2024-01-04 22:06:31 +01:00
if "error" in res:
error_detail = f"Ollama: {res['error']}"
except:
error_detail = f"Ollama: {e}"
2024-01-05 10:25:34 +01:00
raise HTTPException(
2024-01-06 02:16:35 +01:00
status_code=r.status_code if r else 500,
2024-01-05 10:25:34 +01:00
detail=error_detail,
)