feat: add models for gitee.ai (#9490)

api/tests/integration_tests/model_runtime/gitee_ai/test_llm.py

@@ -0,0 +1,132 @@
+import os
+from collections.abc import Generator
+
+import pytest
+
+from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
+from core.model_runtime.entities.message_entities import (
+    AssistantPromptMessage,
+    PromptMessageTool,
+    SystemPromptMessage,
+    UserPromptMessage,
+)
+from core.model_runtime.entities.model_entities import AIModelEntity
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.gitee_ai.llm.llm import GiteeAILargeLanguageModel
+
+
+def test_predefined_models():
+    model = GiteeAILargeLanguageModel()
+    model_schemas = model.predefined_models()
+
+    assert len(model_schemas) >= 1
+    assert isinstance(model_schemas[0], AIModelEntity)
+
+
+def test_validate_credentials_for_chat_model():
+    model = GiteeAILargeLanguageModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        # model name to gpt-3.5-turbo because of mocking
+        model.validate_credentials(model="gpt-3.5-turbo", credentials={"api_key": "invalid_key"})
+
+    model.validate_credentials(
+        model="Qwen2-7B-Instruct",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+    )
+
+
+def test_invoke_chat_model():
+    model = GiteeAILargeLanguageModel()
+
+    result = model.invoke(
+        model="Qwen2-7B-Instruct",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+        prompt_messages=[
+            SystemPromptMessage(
+                content="You are a helpful AI assistant.",
+            ),
+            UserPromptMessage(content="Hello World!"),
+        ],
+        model_parameters={
+            "temperature": 0.0,
+            "top_p": 1.0,
+            "presence_penalty": 0.0,
+            "frequency_penalty": 0.0,
+            "max_tokens": 10,
+            "stream": False,
+        },
+        stop=["How"],
+        stream=False,
+        user="foo",
+    )
+
+    assert isinstance(result, LLMResult)
+    assert len(result.message.content) > 0
+
+
+def test_invoke_stream_chat_model():
+    model = GiteeAILargeLanguageModel()
+
+    result = model.invoke(
+        model="Qwen2-7B-Instruct",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+        prompt_messages=[
+            SystemPromptMessage(
+                content="You are a helpful AI assistant.",
+            ),
+            UserPromptMessage(content="Hello World!"),
+        ],
+        model_parameters={"temperature": 0.0, "max_tokens": 100, "stream": False},
+        stream=True,
+        user="foo",
+    )
+
+    assert isinstance(result, Generator)
+
+    for chunk in result:
+        assert isinstance(chunk, LLMResultChunk)
+        assert isinstance(chunk.delta, LLMResultChunkDelta)
+        assert isinstance(chunk.delta.message, AssistantPromptMessage)
+        assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
+        if chunk.delta.finish_reason is not None:
+            assert chunk.delta.usage is not None
+
+
+def test_get_num_tokens():
+    model = GiteeAILargeLanguageModel()
+
+    num_tokens = model.get_num_tokens(
+        model="Qwen2-7B-Instruct",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+        prompt_messages=[UserPromptMessage(content="Hello World!")],
+    )
+
+    assert num_tokens == 10
+
+    num_tokens = model.get_num_tokens(
+        model="Qwen2-7B-Instruct",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+        prompt_messages=[
+            SystemPromptMessage(
+                content="You are a helpful AI assistant.",
+            ),
+            UserPromptMessage(content="Hello World!"),
+        ],
+        tools=[
+            PromptMessageTool(
+                name="get_weather",
+                description="Determine weather in my location",
+                parameters={
+                    "type": "object",
+                    "properties": {
+                        "location": {"type": "string", "description": "The city and state e.g. San Francisco, CA"},
+                        "unit": {"type": "string", "enum": ["c", "f"]},
+                    },
+                    "required": ["location"],
+                },
+            ),
+        ],
+    )
+
+    assert num_tokens == 77

api/tests/integration_tests/model_runtime/gitee_ai/test_provider.py

@@ -0,0 +1,15 @@
+import os
+
+import pytest
+
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.gitee_ai.gitee_ai import GiteeAIProvider
+
+
+def test_validate_provider_credentials():
+    provider = GiteeAIProvider()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        provider.validate_provider_credentials(credentials={"api_key": "invalid_key"})
+
+    provider.validate_provider_credentials(credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")})

api/tests/integration_tests/model_runtime/gitee_ai/test_rerank.py

@@ -0,0 +1,47 @@
+import os
+
+import pytest
+
+from core.model_runtime.entities.rerank_entities import RerankResult
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.gitee_ai.rerank.rerank import GiteeAIRerankModel
+
+
+def test_validate_credentials():
+    model = GiteeAIRerankModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        model.validate_credentials(
+            model="bge-reranker-v2-m3",
+            credentials={"api_key": "invalid_key"},
+        )
+
+    model.validate_credentials(
+        model="bge-reranker-v2-m3",
+        credentials={
+            "api_key": os.environ.get("GITEE_AI_API_KEY"),
+        },
+    )
+
+
+def test_invoke_model():
+    model = GiteeAIRerankModel()
+    result = model.invoke(
+        model="bge-reranker-v2-m3",
+        credentials={
+            "api_key": os.environ.get("GITEE_AI_API_KEY"),
+        },
+        query="What is the capital of the United States?",
+        docs=[
+            "Carson City is the capital city of the American state of Nevada. At the 2010 United States "
+            "Census, Carson City had a population of 55,274.",
+            "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean that "
+            "are a political division controlled by the United States. Its capital is Saipan.",
+        ],
+        top_n=1,
+        score_threshold=0.01,
+    )
+
+    assert isinstance(result, RerankResult)
+    assert len(result.docs) == 1
+    assert result.docs[0].score >= 0.01

api/tests/integration_tests/model_runtime/gitee_ai/test_speech2text.py

@@ -0,0 +1,45 @@
+import os
+
+import pytest
+
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.gitee_ai.speech2text.speech2text import GiteeAISpeech2TextModel
+
+
+def test_validate_credentials():
+    model = GiteeAISpeech2TextModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        model.validate_credentials(
+            model="whisper-base",
+            credentials={"api_key": "invalid_key"},
+        )
+
+    model.validate_credentials(
+        model="whisper-base",
+        credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")},
+    )
+
+
+def test_invoke_model():
+    model = GiteeAISpeech2TextModel()
+
+    # Get the directory of the current file
+    current_dir = os.path.dirname(os.path.abspath(__file__))
+
+    # Get assets directory
+    assets_dir = os.path.join(os.path.dirname(current_dir), "assets")
+
+    # Construct the path to the audio file
+    audio_file_path = os.path.join(assets_dir, "audio.mp3")
+
+    # Open the file and get the file object
+    with open(audio_file_path, "rb") as audio_file:
+        file = audio_file
+
+        result = model.invoke(
+            model="whisper-base", credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")}, file=file
+        )
+
+        assert isinstance(result, str)
+        assert result == "1 2 3 4 5 6 7 8 9 10"

api/tests/integration_tests/model_runtime/gitee_ai/test_text_embedding.py

@@ -0,0 +1,46 @@
+import os
+
+import pytest
+
+from core.model_runtime.entities.text_embedding_entities import TextEmbeddingResult
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.gitee_ai.text_embedding.text_embedding import GiteeAIEmbeddingModel
+
+
+def test_validate_credentials():
+    model = GiteeAIEmbeddingModel()
+
+    with pytest.raises(CredentialsValidateFailedError):
+        model.validate_credentials(model="bge-large-zh-v1.5", credentials={"api_key": "invalid_key"})
+
+    model.validate_credentials(model="bge-large-zh-v1.5", credentials={"api_key": os.environ.get("GITEE_AI_API_KEY")})
+
+
+def test_invoke_model():
+    model = GiteeAIEmbeddingModel()
+
+    result = model.invoke(
+        model="bge-large-zh-v1.5",
+        credentials={
+            "api_key": os.environ.get("GITEE_AI_API_KEY"),
+        },
+        texts=["hello", "world"],
+        user="user",
+    )
+
+    assert isinstance(result, TextEmbeddingResult)
+    assert len(result.embeddings) == 2
+
+
+def test_get_num_tokens():
+    model = GiteeAIEmbeddingModel()
+
+    num_tokens = model.get_num_tokens(
+        model="bge-large-zh-v1.5",
+        credentials={
+            "api_key": os.environ.get("GITEE_AI_API_KEY"),
+        },
+        texts=["hello", "world"],
+    )
+
+    assert num_tokens == 2

api/tests/integration_tests/model_runtime/gitee_ai/test_tts.py

@@ -0,0 +1,23 @@
+import os
+
+from core.model_runtime.model_providers.gitee_ai.tts.tts import GiteeAIText2SpeechModel
+
+
+def test_invoke_model():
+    model = GiteeAIText2SpeechModel()
+
+    result = model.invoke(
+        model="speecht5_tts",
+        tenant_id="test",
+        credentials={
+            "api_key": os.environ.get("GITEE_AI_API_KEY"),
+        },
+        content_text="Hello, world!",
+        voice="",
+    )
+
+    content = b""
+    for chunk in result:
+        content += chunk
+
+    assert content != b""