stabilität und .env variable
This commit is contained in:
1 parent
fc4c868eaa
commit
16af07dd21
2 files changed
+50
-49
No files matched your search
@@ -1,6 +1,6 @@
|
||||
"""
|
||||
app/services/llm_service.py — LLM Client
|
||||
Version: 2.7.0 (Clean Architecture: Explicit Priority Queues)
|
||||
Version: 2.8.0 (Configurable Concurrency Limit)
|
||||
"""
|
||||
|
||||
import httpx
|
||||
@@ -19,17 +19,28 @@ class Settings:
|
||||
LLM_MODEL = os.getenv("MINDNET_LLM_MODEL", "phi3:mini")
|
||||
PROMPTS_PATH = os.getenv("MINDNET_PROMPTS_PATH", "./config/prompts.yaml")
|
||||
|
||||
# NEU: Konfigurierbares Limit für Hintergrund-Last
|
||||
# Default auf 2 (konservativ), kann in .env erhöht werden.
|
||||
BACKGROUND_LIMIT = int(os.getenv("MINDNET_LLM_BACKGROUND_LIMIT", "2"))
|
||||
|
||||
def get_settings():
|
||||
return Settings()
|
||||
|
||||
class LLMService:
|
||||
# GLOBALER SEMAPHOR (Drosselung für Hintergrund-Prozesse)
|
||||
_background_semaphore = asyncio.Semaphore(2)
|
||||
# GLOBALER SEMAPHOR (Lazy Initialization)
|
||||
# Wir initialisieren ihn erst, wenn wir die Settings kennen.
|
||||
_background_semaphore = None
|
||||
|
||||
def __init__(self):
|
||||
self.settings = get_settings()
|
||||
self.prompts = self._load_prompts()
|
||||
|
||||
# Initialisiere Semaphore einmalig auf Klassen-Ebene basierend auf Config
|
||||
if LLMService._background_semaphore is None:
|
||||
limit = self.settings.BACKGROUND_LIMIT
|
||||
logger.info(f"🚦 LLMService: Initializing Background Semaphore with limit: {limit}")
|
||||
LLMService._background_semaphore = asyncio.Semaphore(limit)
|
||||
|
||||
self.timeout = httpx.Timeout(self.settings.LLM_TIMEOUT, connect=10.0)
|
||||
|
||||
self.client = httpx.AsyncClient(
|
||||
@@ -53,7 +64,7 @@ class LLMService:
|
||||
force_json: bool = False,
|
||||
max_retries: int = 0,
|
||||
base_delay: float = 2.0,
|
||||
priority: Literal["realtime", "background"] = "realtime" # <--- NEU & EXPLIZIT
|
||||
priority: Literal["realtime", "background"] = "realtime"
|
||||
) -> str:
|
||||
"""
|
||||
Führt einen LLM Call aus.
|
||||
@@ -61,13 +72,13 @@ class LLMService:
|
||||
priority="background": Import/Analyse (Gedrosselt durch Semaphore).
|
||||
"""
|
||||
|
||||
# Entscheidung basierend auf explizitem Parameter, nicht Format!
|
||||
use_semaphore = (priority == "background")
|
||||
|
||||
if use_semaphore:
|
||||
if use_semaphore and LLMService._background_semaphore:
|
||||
async with LLMService._background_semaphore:
|
||||
return await self._execute_request(prompt, system, force_json, max_retries, base_delay)
|
||||
else:
|
||||
# Realtime oder Fallback (falls Semaphore Init fehlschlug)
|
||||
return await self._execute_request(prompt, system, force_json, max_retries, base_delay)
|
||||
|
||||
async def _execute_request(self, prompt, system, force_json, max_retries, base_delay):
|
||||
@@ -123,7 +134,7 @@ class LLMService:
|
||||
system=system_prompt,
|
||||
max_retries=0,
|
||||
force_json=False,
|
||||
priority="realtime" # <--- Standard
|
||||
priority="realtime"
|
||||
)
|
||||
|
||||
async def close(self):
|
||||
|
||||
Reference in new issue
Block a user