woody
/
legco_ai_assistant
1
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat: wire LLMClientDP into query decompose pipeline (Phase 6) 

QueryDecomposer now uses LLMClientDP (Deepseek) while RelevanceFilter and RAGService continue using LLMClient (OpenRouter/vLLM).

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
This commit is contained in:
Woody 2026-05-04 14:59:08 +08:00
parent 849beb4d4e
commit 3b5bd79839
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
backend/app/routers/query.py
View File

@ -11,6 +11,7 @@ from app.models.query import QueryRequest, SubQuestionSources
from app.models.common import SourceMetadata
from app.services.history_service import HistoryService
from app.services.llm_client import LLMClient
from app.services.llm_client_dp import LLMClientDP
from app.services.prompt_service import PromptService
from app.services.query_decomposer import QueryDecomposer
from app.services.relevance_filter import RelevanceFilter
@ -177,13 +178,14 @@ async def _query_stream(request: QueryRequest):
try:
history_service = HistoryService(db_path=settings.history_db_path)
llm_client_dp = LLMClientDP(settings)
llm_client = LLMClient(settings)
rag = RAGService(llm_client=llm_client, settings=settings, prompt_service=prompt_service)
active_profile = prompt_service.get_active_profile_name()
logger.info("Query: %s. Active prompt profile: %s", request.question, active_profile)
decomposer = QueryDecomposer(llm_client, prompt_service=prompt_service)
decomposer = QueryDecomposer(llm_client_dp, prompt_service=prompt_service)
# Stage 1: Decompose
stage_start = time.perf_counter()