feat: batch multimodal embeddings

Author: exiaohu

volcenginesdkarkruntime/resources/batch/batch.py

@@ -4,6 +4,7 @@
 from ..._resource import AsyncAPIResource, SyncAPIResource
 from .chat.chat import AsyncChat, Chat
 from .embeddings import AsyncEmbeddings, Embeddings
+from .multimodal_embeddings import AsyncMultimodalEmbeddings, MultimodalEmbeddings
 
 __all__ = ["Batch", "AsyncBatch"]
 
@@ -17,6 +18,10 @@ def chat(self) -> Chat:
     def embeddings(self) -> Embeddings:
         return Embeddings(self._client)
 
+    @cached_property
+    def multimodal_embeddings(self) -> MultimodalEmbeddings:
+        return MultimodalEmbeddings(self._client)
+
 
 class AsyncBatch(AsyncAPIResource):
     @cached_property
@@ -26,3 +31,7 @@ def chat(self) -> AsyncChat:
     @cached_property
     def embeddings(self) -> AsyncEmbeddings:
         return AsyncEmbeddings(self._client)
+
+    @cached_property
+    def multimodal_embeddings(self) -> AsyncMultimodalEmbeddings:
+        return AsyncMultimodalEmbeddings(self._client)

volcenginesdkarkruntime/resources/batch/multimodal_embeddings.py

@@ -0,0 +1,111 @@
+from __future__ import annotations
+
+from typing import List
+
+import httpx
+from typing_extensions import Literal
+
+from ..._base_client import make_request_options
+from ..._compat import cached_property
+from ..._resource import AsyncAPIResource, SyncAPIResource
+from ..._types import Body, Headers, Query
+from ..._utils._utils import async_with_sts_token, with_sts_token
+from ...types.multimodal_embedding import (
+    EmbeddingInputParam,
+    MultimodalEmbeddingResponse,
+)
+from ..multimodal_embeddings import (
+    AsyncMultimodalEmbeddingsWithRawResponse,
+    MultimodalEmbeddingsWithRawResponse,
+)
+from ._utils import async_with_batch_retry, get_request_last_time, with_batch_retry
+
+__all__ = ["MultimodalEmbeddings", "AsyncMultimodalEmbeddings"]
+
+
+class MultimodalEmbeddings(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> MultimodalEmbeddingsWithRawResponse:
+        return MultimodalEmbeddingsWithRawResponse(self)
+
+    @with_sts_token
+    def create(
+        self,
+        *,
+        input: List[EmbeddingInputParam],
+        model: str,
+        encoding_format: Literal["float", "base64"] = "float",
+        dimensions: int | None = None,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None = None,
+    ) -> MultimodalEmbeddingResponse:
+        deadline = get_request_last_time(timeout)
+        breaker = self._client.get_model_breaker(model)
+
+        return with_batch_retry(
+            deadline,
+            breaker,
+            self._post_without_retry,
+            "/batch/embeddings/multimodal",
+            body={
+                "input": input,
+                "model": model,
+                "encoding_format": encoding_format,
+                "dimensions": dimensions,
+            },
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+            ),
+            cast_to=MultimodalEmbeddingResponse,
+        )
+
+
+class AsyncMultimodalEmbeddings(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncMultimodalEmbeddingsWithRawResponse:
+        return AsyncMultimodalEmbeddingsWithRawResponse(self)
+
+    @async_with_sts_token
+    async def create(
+        self,
+        *,
+        input: List[EmbeddingInputParam],
+        model: str,
+        encoding_format: Literal["float", "base64"] = "float",
+        dimensions: int | None = None,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None = None,
+    ) -> MultimodalEmbeddingResponse:
+        deadline = get_request_last_time(timeout)
+        breaker = await self._client.get_model_breaker(model)
+
+        return await async_with_batch_retry(
+            deadline,
+            breaker,
+            self._post_without_retry,
+            "/batch/embeddings/multimodal",
+            body={
+                "input": input,
+                "model": model,
+                "encoding_format": encoding_format,
+                "dimensions": dimensions,
+            },
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+            ),
+            cast_to=MultimodalEmbeddingResponse,
+        )