SDK regeneration

.fern/metadata.json

@@ -25,6 +25,9 @@
         "optional": true
       }
     },
+    "extra_dev_dependencies": {
+      "pydantic": ">=2.0.0,<2.14.0"
+    },
     "improved_imports": true,
     "pydantic_config": {
       "frozen": false,
@@ -100,10 +103,10 @@
       }
     ]
   },
-  "originGitCommit": "2bfdfa67e2f98426451cf784a8f798e6c9272b9b",
+  "originGitCommit": "c39b7388c89804b40d40b481f5338b5a39c93a32",
   "originGitCommitIsDirty": false,
   "invokedBy": "ci",
-  "requestedVersion": "7.2.0",
+  "requestedVersion": "7.3.0",
   "ciProvider": "github",
-  "sdkVersion": "7.2.0"
+  "sdkVersion": "7.3.0"
 }

pyproject.toml

@@ -4,7 +4,7 @@ dynamic = ["version"]
 
 [tool.poetry]
 name = "cohere"
-version = "7.2.0"
+version = "7.3.0"
 description = ""
 readme = "README.md"
 authors = []
@@ -58,6 +58,7 @@ pytest-xdist = "^3.6.1"
 python-dateutil = "^2.9.0"
 types-python-dateutil = "^2.9.0.20240316"
 urllib3 = ">=2.6.3,<3.0.0"
+pydantic = ">=2.0.0,<2.14.0"
 ruff = "==0.11.5"
 
 [tool.pytest.ini_options]

reference.md

@@ -1510,8 +1510,8 @@ client.embed(
         "hello",
         "goodbye"
     ],
-    model="embed-v4.0",
-    input_type="classification",
+    model="embed-v5.0-fast",
+    input_type="search_document",
 )
 
 ```
@@ -3025,8 +3025,8 @@ client.v2.embed(
         "hello",
         "goodbye"
     ],
-    model="embed-v4.0",
-    input_type="classification",
+    model="embed-v5.0-fast",
+    input_type="search_document",
     embedding_types=[
         "float"
     ],
@@ -3107,7 +3107,7 @@ For **Embed v4.0 and newer** models, there is no limit on the number of images p
 **output_dimension:** `typing.Optional[int]` 
 
 The number of dimensions of the output embedding. This is only available for `embed-v4` and newer models.
-Possible values are `256`, `512`, `1024`, and `1536`. The default is `1536`.
+For supported dimensions, please refer to the [Embed models](https://docs.cohere.com/docs/cohere-embed) page.
 
 </dd>
 </dl>
@@ -3193,7 +3193,7 @@ This endpoint takes in a query and a list of texts and produces an ordered array
 <dd>
 
 ```python
-from cohere import Client
+from cohere import Client, RerankRelevantSpans
 from cohere.environment import ClientEnvironment
 
 client = Client(
@@ -3202,16 +3202,21 @@ client = Client(
 )
 
 client.v2.rerank(
-    model="rerank-v4.0-pro",
+    model="rerank-v5.0-pro",
     query="What is the capital of the United States?",
     documents=[
         "Carson City is the capital city of the American state of Nevada.",
         "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean. Its capital is Saipan.",
         "Capitalization or capitalisation in English grammar is the use of a capital letter at the start of a word. English usage varies from capitalization in other languages.",
         "Washington, D.C. (also known as simply Washington or D.C., and officially as the District of Columbia) is the capital of the United States. It is a federal district.",
-        "Capital punishment has existed in the United States since beforethe United States was a country. As of 2017, capital punishment is legal in 30 of the 50 states."
+        "Capital gains are profits from the sale of an asset such as stocks or real estate. In the United States, long-term capital gains are taxed at lower rates than ordinary income."
     ],
     top_n=3,
+    relevant_spans=RerankRelevantSpans(
+        mode="enabled",
+        threshold=0.1,
+        max_spans=3,
+    ),
 )
 
 ```
@@ -3228,7 +3233,7 @@ client.v2.rerank(
 <dl>
 <dd>
 
-**model:** `str` — The identifier of the model to use, eg `rerank-v3.5`.
+**model:** `str` — The identifier of the model to use, eg `rerank-v5.0-pro`.
 
 </dd>
 </dl>
@@ -3275,6 +3280,14 @@ For optimal performance we recommend against sending more than 1,000 documents i
 <dl>
 <dd>
 
+**relevant_spans:** `typing.Optional[RerankRelevantSpans]` 
+
+</dd>
+</dl>
+
+<dl>
+<dd>
+
 **priority:** `typing.Optional[int]` — Controls how early the request is handled. Lower numbers indicate higher priority (default: 0, the highest). When the system is under load, higher-priority requests are processed first and are the least likely to be dropped.
 
 </dd>

src/cohere/__init__.py

@@ -196,6 +196,9 @@
         ParseTextBlock,
         ParseTextContentBlock,
         RerankDocument,
+        RerankRelevantSpanResponse,
+        RerankRelevantSpans,
+        RerankRelevantSpansMode,
         RerankRequestDocumentsItem,
         RerankResponse,
         RerankResponseResultsItem,
@@ -572,6 +575,9 @@
     "ParseTextBlock": ".types",
     "ParseTextContentBlock": ".types",
     "RerankDocument": ".types",
+    "RerankRelevantSpanResponse": ".types",
+    "RerankRelevantSpans": ".types",
+    "RerankRelevantSpansMode": ".types",
     "RerankRequestDocumentsItem": ".types",
     "RerankResponse": ".types",
     "RerankResponseResultsItem": ".types",
@@ -934,6 +940,9 @@ def __dir__():
     "ParseTextBlock",
     "ParseTextContentBlock",
     "RerankDocument",
+    "RerankRelevantSpanResponse",
+    "RerankRelevantSpans",
+    "RerankRelevantSpansMode",
     "RerankRequestDocumentsItem",
     "RerankResponse",
     "RerankResponseResultsItem",

src/cohere/base_client.py

@@ -1147,8 +1147,8 @@ def embed(
         )
         client.embed(
             texts=["hello", "goodbye"],
-            model="embed-v4.0",
-            input_type="classification",
+            model="embed-v5.0-fast",
+            input_type="search_document",
         )
         """
         _response = self._raw_client.embed(
@@ -2747,8 +2747,8 @@ async def embed(
         async def main() -> None:
             await client.embed(
                 texts=["hello", "goodbye"],
-                model="embed-v4.0",
-                input_type="classification",
+                model="embed-v5.0-fast",
+                input_type="search_document",
             )
 
 

src/cohere/core/client_wrapper.py

@@ -35,12 +35,12 @@ def get_headers(self) -> typing.Dict[str, str]:
         import platform
 
         headers: typing.Dict[str, str] = {
-            "User-Agent": "cohere/7.2.0",
+            "User-Agent": "cohere/7.3.0",
             "X-Fern-Language": "Python",
             "X-Fern-Runtime": f"python/{platform.python_version()}",
             "X-Fern-Platform": f"{platform.system().lower()}/{platform.release()}",
             "X-Fern-SDK-Name": "cohere",
-            "X-Fern-SDK-Version": "7.2.0",
+            "X-Fern-SDK-Version": "7.3.0",
             **(self.get_custom_headers() or {}),
         }
         if self._client_name is not None:

src/cohere/types/__init__.py

@@ -203,6 +203,9 @@
     from .parse_text_block import ParseTextBlock
     from .parse_text_content_block import ParseTextContentBlock
     from .rerank_document import RerankDocument
+    from .rerank_relevant_span_response import RerankRelevantSpanResponse
+    from .rerank_relevant_spans import RerankRelevantSpans
+    from .rerank_relevant_spans_mode import RerankRelevantSpansMode
     from .rerank_request_documents_item import RerankRequestDocumentsItem
     from .rerank_response import RerankResponse
     from .rerank_response_results_item import RerankResponseResultsItem
@@ -444,6 +447,9 @@
     "ParseTextBlock": ".parse_text_block",
     "ParseTextContentBlock": ".parse_text_content_block",
     "RerankDocument": ".rerank_document",
+    "RerankRelevantSpanResponse": ".rerank_relevant_span_response",
+    "RerankRelevantSpans": ".rerank_relevant_spans",
+    "RerankRelevantSpansMode": ".rerank_relevant_spans_mode",
     "RerankRequestDocumentsItem": ".rerank_request_documents_item",
     "RerankResponse": ".rerank_response",
     "RerankResponseResultsItem": ".rerank_response_results_item",
@@ -727,6 +733,9 @@ def __dir__():
     "ParseTextBlock",
     "ParseTextContentBlock",
     "RerankDocument",
+    "RerankRelevantSpanResponse",
+    "RerankRelevantSpans",
+    "RerankRelevantSpansMode",
     "RerankRequestDocumentsItem",
     "RerankResponse",
     "RerankResponseResultsItem",

src/cohere/types/rerank_relevant_span_response.py

@@ -0,0 +1,36 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import pydantic
+from ..core.pydantic_utilities import IS_PYDANTIC_V2
+from ..core.unchecked_base_model import UncheckedBaseModel
+
+
+class RerankRelevantSpanResponse(UncheckedBaseModel):
+    """
+    A stretch of a document that the model found relevant to the query, returned when context compression is enabled through `relevant_spans`.
+    """
+
+    start: int = pydantic.Field()
+    """
+    The position of the span's first character in the document, counted in Unicode codepoints.
+    """
+
+    end: int = pydantic.Field()
+    """
+    The position just past the span's last character, counted in Unicode codepoints, so the span covers `[start, end)` of the document.
+    """
+
+    score: float = pydantic.Field()
+    """
+    How relevant the span is to the query, in the range `[0, 1]`. Only spans reaching the requested `threshold` are returned.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow")  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            smart_union = True
+            extra = pydantic.Extra.allow

src/cohere/types/rerank_relevant_spans.py

@@ -0,0 +1,37 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import pydantic
+from ..core.pydantic_utilities import IS_PYDANTIC_V2
+from ..core.unchecked_base_model import UncheckedBaseModel
+from .rerank_relevant_spans_mode import RerankRelevantSpansMode
+
+
+class RerankRelevantSpans(UncheckedBaseModel):
+    """
+    When enabled, the API response will include the relevant spans of each document that contributed to its rank. This feature is only supported for Rerank v5 and newer models.
+    """
+
+    mode: RerankRelevantSpansMode = pydantic.Field()
+    """
+    One of `"enabled"` or `"disabled"`. Controls whether relevant spans will be returned in the API response.
+    """
+
+    threshold: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    Only spans with a relevance score higher than the specified threshold value will be included in the response. Defaults to `0.1`.
+    """
+
+    max_spans: typing.Optional[int] = pydantic.Field(default=None)
+    """
+    The maximum number of relevant spans to include per document. If not set, all relevant spans will be returned.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow")  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            smart_union = True
+            extra = pydantic.Extra.allow

src/cohere/types/rerank_relevant_spans_mode.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+RerankRelevantSpansMode = typing.Union[typing.Literal["enabled", "disabled"], typing.Any]