mirror of
https://github.com/infiniflow/ragflow.git
synced 2026-05-29 20:17:35 +08:00
3f02ca7ba1
### What problem does this PR solve?
implement embed, rerank, tts for AstraFlow
**Verify from CLI**
```
# Astraflow
RAGFlow(user)> tts with 'IndexTeam/IndexTTS-2@test3@astraflow' text 'hello? show yourself' play format 'wav' param '{"voice": "jack_cheng"}'
SUCCESS
RAGFlow(user)> rerank query 'what is rag' document 'rag is retrieval augment generation' 'rag need llm' 'famous rag project includes ragflow' with 'bge-reranker-v2-m3@test3@astraflow' top 3;
+-------+---------------------+
| index | relevance_score |
+-------+---------------------+
| 0 | 0.9837390184402466 |
| 2 | 0.06322699040174484 |
| 1 | 0.04663187265396118 |
+-------+---------------------+
RAGFlow(user)> embed text 'walkerwhat' 'jumperwho' with 'text-embedding-3-large@test3@astraflow' dimension 16
+-----------+-------+
| dimension | index |
+-----------+-------+
| 3072 | 0 |
| 3072 | 1 |
+-----------+-------+
# Xinference
```
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- [x] Refactoring
163 lines
2.8 KiB
JSON
163 lines
2.8 KiB
JSON
{
|
|
"name": "Astraflow",
|
|
"url": {
|
|
"default": "https://api.modelverse.cn/v1",
|
|
"us-ca": "https://api-us-ca.umodelverse.ai/v1"
|
|
},
|
|
"url_suffix": {
|
|
"chat": "chat/completions",
|
|
"models": "models",
|
|
"embedding": "embeddings",
|
|
"rerank": "rerank",
|
|
"tts": "audio/speech"
|
|
},
|
|
"class": "astraflow",
|
|
"models": [
|
|
{
|
|
"name": "text-embedding-3-large",
|
|
"max_tokens": 16384,
|
|
"model_types": [
|
|
"embedding"
|
|
]
|
|
},
|
|
{
|
|
"name": "bge-reranker-v2-m3",
|
|
"max_tokens": 8192,
|
|
"model_types": [
|
|
"rerank"
|
|
]
|
|
},
|
|
{
|
|
"name": "IndexTeam/IndexTTS-2",
|
|
"model_types": [
|
|
"tts"
|
|
]
|
|
},
|
|
{
|
|
"name": "claude-opus-4-7",
|
|
"max_tokens": 200000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "claude-opus-4-6",
|
|
"max_tokens": 200000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "claude-sonnet-4-5-20250929",
|
|
"max_tokens": 200000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "claude-haiku-4-5-20251001",
|
|
"max_tokens": 200000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gpt-5.4",
|
|
"max_tokens": 400000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gpt-5.4-mini",
|
|
"max_tokens": 400000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gpt-5.4-nano",
|
|
"max_tokens": 400000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gpt-4o-mini",
|
|
"max_tokens": 128000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "Qwen/Qwen3-Max",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "Qwen/Qwen3-Coder",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "Qwen/Qwen3-32B",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "Qwen/Qwen3-VL-235B-A22B-Instruct",
|
|
"max_tokens": 131072,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "kimi-k2.6",
|
|
"max_tokens": 200000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "glm-5.1",
|
|
"max_tokens": 128000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "MiniMax-M2.7",
|
|
"max_tokens": 1000000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "MiniMax-M2",
|
|
"max_tokens": 1000000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gemini-2.5-pro",
|
|
"max_tokens": 1000000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
},
|
|
{
|
|
"name": "gemini-2.5-flash",
|
|
"max_tokens": 1000000,
|
|
"model_types": [
|
|
"chat"
|
|
]
|
|
}
|
|
]
|
|
} |