gmakstutis/ragflow
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix: Asure-OpenAI resource not found

Browse source
This commit is contained in:
yongtenglei 2025-12-12 19:02:37 +08:00
parent 948bc93786
commit 419efaa395
2 changed files with 14 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
rag/llm/__init__.py
View file

@ -55,6 +55,7 @@ class SupportedLiteLLMProvider(StrEnum):
DeerAPI = "DeerAPI" DeerAPI = "DeerAPI"
GPUStack = "GPUStack" GPUStack = "GPUStack"
OpenAI = "OpenAI" OpenAI = "OpenAI"
Azure_OpenAI = "Azure-OpenAI"
FACTORY_DEFAULT_BASE_URL = { FACTORY_DEFAULT_BASE_URL = {
@ -116,7 +117,7 @@ LITELLM_PROVIDER_PREFIX = {
SupportedLiteLLMProvider.DeerAPI: "openai/", SupportedLiteLLMProvider.DeerAPI: "openai/",
SupportedLiteLLMProvider.GPUStack: "openai/", SupportedLiteLLMProvider.GPUStack: "openai/",
SupportedLiteLLMProvider.OpenAI: "openai/", SupportedLiteLLMProvider.OpenAI: "openai/",
SupportedLiteLLMProvider.Azure_OpenAI: "azure/",
} }
ChatModel = globals().get("ChatModel", {}) ChatModel = globals().get("ChatModel", {})

13
rag/llm/chat_model.py
View file

@ -191,7 +191,7 @@ class Base(ABC):
except Exception as e: except Exception as e:
e = await self._exceptions_async(e, attempt) e = await self._exceptions_async(e, attempt)
if e: if e:
yield e yield e
yield total_tokens yield total_tokens
return return
@ -1219,6 +1219,7 @@ class LiteLLMBase(ABC):
"DeerAPI", "DeerAPI",
"GPUStack", "GPUStack",
"OpenAI", "OpenAI",
"Azure-OpenAI",
] ]
def __init__(self, key, model_name, base_url=None, **kwargs): def __init__(self, key, model_name, base_url=None, **kwargs):
@ -1244,6 +1245,9 @@ class LiteLLMBase(ABC):
elif self.provider == SupportedLiteLLMProvider.OpenRouter: elif self.provider == SupportedLiteLLMProvider.OpenRouter:
self.api_key = json.loads(key).get("api_key", "") self.api_key = json.loads(key).get("api_key", "")
self.provider_order = json.loads(key).get("provider_order", "") self.provider_order = json.loads(key).get("provider_order", "")
elif self.provider == SupportedLiteLLMProvider.Azure_OpenAI:
self.api_key = json.loads(key).get("api_key", "")
self.api_version = json.loads(key).get("api_version", "2024-02-01")
def _get_delay(self): def _get_delay(self):
return self.base_delay * random.uniform(10, 150) return self.base_delay * random.uniform(10, 150)
@ -1675,6 +1679,13 @@ class LiteLLMBase(ABC):
"api_base": self.base_url, "api_base": self.base_url,
} }
) )
elif self.provider == SupportedLiteLLMProvider.GPUStack:
completion_args.update(
{
"api_base": self.base_url,
"api_version": self.api_version,
}
)
# Ollama deployments commonly sit behind a reverse proxy that enforces # Ollama deployments commonly sit behind a reverse proxy that enforces
# Bearer auth. Ensure the Authorization header is set when an API key # Bearer auth. Ensure the Authorization header is set when an API key