zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-06-29 23:41:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat: Add Astraflow provider support (global + China endpoints) (#14270)

Browse Source 
## Add Astraflow Provider Support

This PR integrates [Astraflow](https://astraflow.ucloud.cn/) (by UCloud
/ 优刻得) as a new AI model provider in RAGFlow, with support for both
global and China endpoints.

### About Astraflow
Astraflow is an OpenAI-compatible AI model aggregation platform
supporting 200+ models from major providers including DeepSeek, Qwen,
GPT, Claude, Gemini, Llama, Mistral, and more.

| Variant | Factory Name | Endpoint | Env Var |
|---------|-------------|----------|---------|
| Global | `Astraflow` | `https://api-us-ca.umodelverse.ai/v1` |
`ASTRAFLOW_API_KEY` |
| China | `Astraflow-CN` | `https://api.modelverse.cn/v1` |
`ASTRAFLOW_CN_API_KEY` |

- **API key signup**: https://astraflow.ucloud.cn/

---

### Files Changed

| File | Change |
|------|--------|
| `rag/llm/__init__.py` | Register `Astraflow` and `Astraflow-CN` in
`SupportedLiteLLMProvider` enum, `FACTORY_DEFAULT_BASE_URL`, and
`LITELLM_PROVIDER_PREFIX` |
| `rag/llm/chat_model.py` | Add `AstraflowChat` and `AstraflowCNChat`
(OpenAI-compatible `Base` subclass) |
| `rag/llm/embedding_model.py` | Add `AstraflowEmbed` and
`AstraflowCNEmbed` (subclasses of `OpenAIEmbed`) |
| `rag/llm/rerank_model.py` | Add `AstraflowRerank` and
`AstraflowCNRerank` (subclasses of `OpenAI_APIRerank`) |
| `rag/llm/cv_model.py` | Add `AstraflowCV` and `AstraflowCNCV`
(subclasses of `GptV4`) |
| `rag/llm/tts_model.py` | Add `AstraflowTTS` and `AstraflowCNTTS`
(subclasses of `OpenAITTS`) |
| `rag/llm/sequence2txt_model.py` | Add `AstraflowSeq2txt` and
`AstraflowCNSeq2txt` (subclasses of `GPTSeq2txt`) |
| `conf/llm_factories.json` | Register `Astraflow` and `Astraflow-CN`
factories with a curated list of popular models |

---

### Supported Model Types
- ✅ **Chat / LLM** — DeepSeek-V3/R1, Qwen3, GPT-4o/4.1, Claude 3.5/3.7,
Gemini 2.0/2.5 Flash, Llama 3.3/4, Mistral, and 200+ more
- ✅ **Text Embedding** — text-embedding-3-small/large
- ✅ **Image / Vision (IMAGE2TEXT)** — GPT-4o, GPT-4.1, Claude, Gemini,
Llama-4, etc.
- ✅ **Text Re-Rank**
- ✅ **TTS** — tts-1
- ✅ **Speech-to-Text (SPEECH2TEXT)** — whisper-1

### Implementation Notes
- Uses the `openai/` LiteLLM prefix — consistent with other
OpenAI-compatible aggregation platforms (SILICONFLOW, DeerAPI, CometAPI,
OpenRouter, n1n, Avian, etc.)
- `Astraflow` (global, rank 250) and `Astraflow-CN` (China, rank 249)
are separate factory entries, allowing users to choose the optimal
endpoint based on their region.
- All model classes cleanly subclass existing base classes (`Base`,
`OpenAIEmbed`, `OpenAI_APIRerank`, `GptV4`, `OpenAITTS`, `GPTSeq2txt`)
with no custom logic needed — the provider is fully OpenAI-compatible.

---------

Co-authored-by: user <user@xzaaaMacBook-Air.local>
This commit is contained in:
ucloudnb666
2026-04-22 15:38:34 +08:00
committed by GitHub
parent d5d162b374
commit f853a39b40
5 changed files with 396 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

387
conf/llm_factories.json
View File

@@ -377,7 +377,7 @@
"tags": "LLM,TEXT EMBEDDING,TEXT RE-RANK,TTS,SPEECH2TEXT,MODERATION",
"status": "1",
"rank": "950",
"url" : "https://dashscope.aliyuncs.com/compatible-mode/v1",
"url": "https://dashscope.aliyuncs.com/compatible-mode/v1",
"llm": [
{
"llm_name": "qwen3.5-122b-a10b",
@@ -1557,53 +1557,52 @@
"rank": "980",
"llm": [
{
"llm_name": "gemini-3-pro-preview",
"tags": "LLM,CHAT,1M,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-3-pro-preview",
"tags": "LLM,CHAT,1M,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-2.5-flash",
"tags": "LLM,CHAT,1024K,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-2.5-flash",
"tags": "LLM,CHAT,1024K,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-2.5-pro",
"tags": "LLM,CHAT,IMAGE2TEXT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-2.5-pro",
"tags": "LLM,CHAT,IMAGE2TEXT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-2.5-flash-lite",
"tags": "LLM,CHAT,1024K,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-2.5-flash-lite",
"tags": "LLM,CHAT,1024K,IMAGE2TEXT",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-2.0-flash",
"tags": "LLM,CHAT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-2.0-flash",
"tags": "LLM,CHAT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-2.0-flash-lite",
"tags": "LLM,CHAT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
"llm_name": "gemini-2.0-flash-lite",
"tags": "LLM,CHAT,1024K",
"max_tokens": 1048576,
"model_type": "image2text",
"is_tools": true
},
{
"llm_name": "gemini-embedding-001",
"tags": "TEXT EMBEDDING",
"max_tokens": 2048,
"model_type": "embedding"
"llm_name": "gemini-embedding-001",
"tags": "TEXT EMBEDDING",
"max_tokens": 2048,
"model_type": "embedding"
}
]
},
@@ -6293,6 +6292,320 @@
}
]
},
{
"name": "Astraflow",
"logo": "",
"tags": "LLM,TEXT EMBEDDING",
"status": "1",
"rank": "250",
"url": "https://api-us-ca.umodelverse.ai/v1",
"llm": [
{
"llm_name": "claude-opus-4-7",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-opus-4-6",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-sonnet-4-5-20250929",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-haiku-4-5-20251001",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4-mini",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4-nano",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-4o-mini",
"tags": "LLM,CHAT,128k",
"max_tokens": 128000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-Max",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-Coder",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-32B",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-VL-235B-A22B-Instruct",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "kimi-k2.6",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "glm-5.1",
"tags": "LLM,CHAT,128k",
"max_tokens": 128000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "MiniMax-M2.7",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "MiniMax-M2",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gemini-2.5-pro",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gemini-2.5-flash",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "qwen3-embedding-8b",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8192,
"model_type": "embedding",
"is_tools": false
},
{
"llm_name": "text-embedding-3-large",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8191,
"model_type": "embedding",
"is_tools": false
},
{
"llm_name": "text-embedding-ada-002",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8191,
"model_type": "embedding",
"is_tools": false
}
]
},
{
"name": "Astraflow-CN",
"logo": "",
"tags": "LLM,TEXT EMBEDDING",
"status": "1",
"rank": "249",
"url": "https://api.modelverse.cn/v1",
"llm": [
{
"llm_name": "claude-opus-4-7",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-opus-4-6",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-sonnet-4-5-20250929",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "claude-haiku-4-5-20251001",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4-mini",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-5.4-nano",
"tags": "LLM,CHAT,400k",
"max_tokens": 400000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gpt-4o-mini",
"tags": "LLM,CHAT,128k",
"max_tokens": 128000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-Max",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-Coder",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-32B",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "Qwen/Qwen3-VL-235B-A22B-Instruct",
"tags": "LLM,CHAT,131k",
"max_tokens": 131072,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "kimi-k2.6",
"tags": "LLM,CHAT,200k",
"max_tokens": 200000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "glm-5.1",
"tags": "LLM,CHAT,128k",
"max_tokens": 128000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "MiniMax-M2.7",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "MiniMax-M2",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gemini-2.5-pro",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "gemini-2.5-flash",
"tags": "LLM,CHAT,1M",
"max_tokens": 1000000,
"model_type": "chat",
"is_tools": true
},
{
"llm_name": "qwen3-embedding-8b",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8192,
"model_type": "embedding",
"is_tools": false
},
{
"llm_name": "text-embedding-3-large",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8191,
"model_type": "embedding",
"is_tools": false
},
{
"llm_name": "text-embedding-ada-002",
"tags": "TEXT EMBEDDING,8K",
"max_tokens": 8191,
"model_type": "embedding",
"is_tools": false
}
]
},
{
"name": "Avian",
"logo": "",
@@ -6370,4 +6683,4 @@
]
}
]
}
}

2
docs/guides/models/supported_models.mdx
View File

@@ -18,6 +18,8 @@ A complete list of models supported by RAGFlow, which will continue to expand.
| Provider | LLM | Image2Text | Speech2text | TTS | Embedding | Rerank | OCR |
| --------------------- | ------------------ | ------------------ | ------------------ | ------------------ | ------------------ | ------------------ | ------------------ |
| Anthropic | :heavy_check_mark: | | | | | | |
| Astraflow | :heavy_check_mark: | | | | :heavy_check_mark: | | |
| Astraflow-CN | :heavy_check_mark: | | | | :heavy_check_mark: | | |
| Avian | :heavy_check_mark: | | | | | | |
| Azure-OpenAI | :heavy_check_mark: | :heavy_check_mark: | :heavy_check_mark: | | :heavy_check_mark: | | |
| BaiChuan | :heavy_check_mark: | | | | :heavy_check_mark: | | |

6
rag/llm/__init__.py
View File

@@ -59,6 +59,8 @@ class SupportedLiteLLMProvider(StrEnum):
n1n = "n1n"
HunYuan = "Tencent Hunyuan"
Avian = "Avian"
Astraflow = "Astraflow"
Astraflow_CN = "Astraflow-CN"
FACTORY_DEFAULT_BASE_URL = {
@@ -87,6 +89,8 @@ FACTORY_DEFAULT_BASE_URL = {
SupportedLiteLLMProvider.n1n: "https://api.n1n.ai/v1",
SupportedLiteLLMProvider.HunYuan: "https://api.hunyuan.cloud.tencent.com/v1",
SupportedLiteLLMProvider.Avian: "https://api.avian.io/v1",
SupportedLiteLLMProvider.Astraflow: "https://api-us-ca.umodelverse.ai/v1",
SupportedLiteLLMProvider.Astraflow_CN: "https://api.modelverse.cn/v1",
}
@@ -127,6 +131,8 @@ LITELLM_PROVIDER_PREFIX = {
SupportedLiteLLMProvider.n1n: "openai/",
SupportedLiteLLMProvider.HunYuan: "openai/",
SupportedLiteLLMProvider.Avian: "openai/",
SupportedLiteLLMProvider.Astraflow: "openai/",
SupportedLiteLLMProvider.Astraflow_CN: "openai/",
}
ChatModel = globals().get("ChatModel", {})

21
rag/llm/chat_model.py
View File

@@ -30,11 +30,12 @@ import openai
from openai import AsyncOpenAI, OpenAI
from strenum import StrEnum
from common.misc_utils import thread_pool_exec
from common.token_utils import num_tokens_from_string, total_token_count_from_response
from rag.llm import FACTORY_DEFAULT_BASE_URL, LITELLM_PROVIDER_PREFIX, SupportedLiteLLMProvider
from rag.nlp import is_chinese, is_english
from common.misc_utils import thread_pool_exec
class LLMErrorCode(StrEnum):
ERROR_RATE_LIMIT = "RATE_LIMIT_EXCEEDED"
ERROR_AUTHENTICATION = "AUTH_ERROR"
@@ -1208,6 +1209,24 @@ class AvianChat(Base):
super().__init__(key, model_name, base_url, **kwargs)
class AstraflowChat(Base):
_FACTORY_NAME = "Astraflow"
def __init__(self, key, model_name, base_url="https://api-us-ca.umodelverse.ai/v1", **kwargs):
if not base_url:
base_url = "https://api-us-ca.umodelverse.ai/v1"
super().__init__(key, model_name, base_url, **kwargs)
class AstraflowCNChat(Base):
_FACTORY_NAME = "Astraflow-CN"
def __init__(self, key, model_name, base_url="https://api.modelverse.cn/v1", **kwargs):
if not base_url:
base_url = "https://api.modelverse.cn/v1"
super().__init__(key, model_name, base_url, **kwargs)
class LiteLLMBase(ABC):
_FACTORY_NAME = [
"Tongyi-Qianwen",

18
rag/llm/embedding_model.py
View File

@@ -161,6 +161,24 @@ class AzureEmbed(OpenAIEmbed):
self.model_name = model_name
class AstraflowEmbed(OpenAIEmbed):
_FACTORY_NAME = "Astraflow"
def __init__(self, key, model_name, base_url="https://api-us-ca.umodelverse.ai/v1"):
if not base_url:
base_url = "https://api-us-ca.umodelverse.ai/v1"
super().__init__(key, model_name, base_url)
class AstraflowCNEmbed(OpenAIEmbed):
_FACTORY_NAME = "Astraflow-CN"
def __init__(self, key, model_name, base_url="https://api.modelverse.cn/v1"):
if not base_url:
base_url = "https://api.modelverse.cn/v1"
super().__init__(key, model_name, base_url)
class BaiChuanEmbed(OpenAIEmbed):
_FACTORY_NAME = "BaiChuan"