70 lines
2.2 KiB
Python
70 lines
2.2 KiB
Python
import logging
|
|
from typing import Dict
|
|
|
|
import httpx
|
|
from config.setting import OLLAMA_URL
|
|
from fastapi.responses import JSONResponse
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
class ModelInfoService:
|
|
OUTER_MODELS = [
|
|
"claude-sonnet-4-20250514",
|
|
"claude-3-7-sonnet-20250219",
|
|
"claude-3-5-haiku-20241022",
|
|
"gemini-2.5-pro",
|
|
"gemini-2.5-flash",
|
|
"gpt-4.1",
|
|
"gpt-4o",
|
|
]
|
|
|
|
@staticmethod
|
|
async def get_ollama_model_map() -> Dict[str, str]:
|
|
model_url_map = {}
|
|
for url in OLLAMA_URL:
|
|
try:
|
|
async with httpx.AsyncClient(timeout=3.0) as client:
|
|
tags_url = url.replace("/api/generate", "/api/tags")
|
|
res = await client.get(tags_url)
|
|
res.raise_for_status()
|
|
models = res.json().get("models", [])
|
|
for m in models:
|
|
model_url_map[m["name"]] = url
|
|
except Exception as e:
|
|
logger.error(f"[ERROR] {url} 모델 조회 실패: {e}")
|
|
return model_url_map
|
|
|
|
@staticmethod
|
|
async def get_model_info() -> JSONResponse:
|
|
inner_models = []
|
|
|
|
for url in OLLAMA_URL:
|
|
try:
|
|
async with httpx.AsyncClient(timeout=3.0) as client:
|
|
tags_url = url.replace("/generate", "/tags")
|
|
res = await client.get(tags_url)
|
|
res.raise_for_status()
|
|
data = res.json()
|
|
models = [m["name"] for m in data.get("models", [])]
|
|
inner_models.extend(models)
|
|
except Exception as e:
|
|
logger.error(f"[API-INFO-ERROR] Ollama 모델 조회 실패 ({url}): {e}")
|
|
|
|
inner_models = list(set(inner_models))
|
|
|
|
return JSONResponse(
|
|
content={
|
|
"models": {
|
|
"inner_model": {
|
|
"default_model": "gpt-oss:20b", # gemma3:27b
|
|
"model_list": inner_models,
|
|
},
|
|
"outer_model": {
|
|
"default_model": "gpt-4.1",
|
|
"model_list": ModelInfoService.OUTER_MODELS,
|
|
},
|
|
}
|
|
}
|
|
)
|