apunkt/SurfSense
1
0
Fork 0
mirror of https://github.com/MODSetter/SurfSense.git synced 2026-05-17 18:35:19 +02:00
Code Issues Projects Releases Packages Wiki Activity

fix: Added API_BASE param for LiteLLM.

Browse source
This commit is contained in:
DESKTOP-RTLN3BA\$punk 2025-05-08 19:30:56 -07:00
parent 9f3d49ab93
commit 934aff2518
7 changed files with 151 additions and 73 deletions
Download patch file Download diff file Expand all files Collapse all files

5
surfsense_backend/.env.example
View file

@ -30,3 +30,8 @@ LANGSMITH_TRACING=true
LANGSMITH_ENDPOINT="https://api.smith.langchain.com"
LANGSMITH_API_KEY="lsv2_pt_....."
LANGSMITH_PROJECT="surfsense"
# OPTIONAL: LiteLLM API Base
FAST_LLM_API_BASE=""
STRATEGIC_LLM_API_BASE=""
LONG_CONTEXT_LLM_API_BASE=""

1
surfsense_backend/.gitignore vendored
View file

@ -5,3 +5,4 @@ data/
__pycache__/
.flashrank_cache
surf_new_backend.egg-info/
podcasts/

4
surfsense_backend/app/agents/researcher/state.py
View file

@ -3,7 +3,7 @@
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Optional, Any
from typing import List, Optional, Any
from sqlalchemy.ext.asyncio import AsyncSession
from app.utils.streaming_service import StreamingService
@ -21,7 +21,7 @@ class State:
# Streaming service
streaming_service: StreamingService
# Intermediate state - populated during workflow
# chat_history: Optional[List[Any]] = field(default=None)
# Using field to explicitly mark as part of state
answer_outline: Optional[Any] = field(default=None)

23
surfsense_backend/app/config/__init__.py
View file

@ -46,14 +46,27 @@ class Config:
# LONG-CONTEXT LLMS
LONG_CONTEXT_LLM = os.getenv("LONG_CONTEXT_LLM")
long_context_llm_instance = ChatLiteLLM(model=LONG_CONTEXT_LLM)
FAST_LLM_API_BASE = os.getenv("FAST_LLM_API_BASE")
if FAST_LLM_API_BASE:
long_context_llm_instance = ChatLiteLLM(model=LONG_CONTEXT_LLM, api_base=FAST_LLM_API_BASE)
else:
long_context_llm_instance = ChatLiteLLM(model=LONG_CONTEXT_LLM)
# FAST & STRATEGIC LLM's
# FAST LLM
FAST_LLM = os.getenv("FAST_LLM")
STRATEGIC_LLM = os.getenv("STRATEGIC_LLM")
fast_llm_instance = ChatLiteLLM(model=FAST_LLM)
strategic_llm_instance = ChatLiteLLM(model=STRATEGIC_LLM)
FAST_LLM_API_BASE = os.getenv("FAST_LLM_API_BASE")
if FAST_LLM_API_BASE:
fast_llm_instance = ChatLiteLLM(model=FAST_LLM, api_base=FAST_LLM_API_BASE)
else:
fast_llm_instance = ChatLiteLLM(model=FAST_LLM)
# STRATEGIC LLM
STRATEGIC_LLM = os.getenv("STRATEGIC_LLM")
STRATEGIC_LLM_API_BASE = os.getenv("STRATEGIC_LLM_API_BASE")
if STRATEGIC_LLM_API_BASE:
strategic_llm_instance = ChatLiteLLM(model=STRATEGIC_LLM, api_base=STRATEGIC_LLM_API_BASE)
else:
strategic_llm_instance = ChatLiteLLM(model=STRATEGIC_LLM)
# Chonkie Configuration | Edit this to your needs
EMBEDDING_MODEL = os.getenv("EMBEDDING_MODEL")