SDK regeneration

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "cohere"
-version = "5.0.0a11"
+version = "5.0.0a12"
 description = ""
 readme = "README.md"
 authors = []

src/cohere/base_client.py

@@ -128,6 +128,7 @@ def chat_stream(
         max_tokens: typing.Optional[int] = OMIT,
         k: typing.Optional[int] = OMIT,
         p: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         frequency_penalty: typing.Optional[float] = OMIT,
         presence_penalty: typing.Optional[float] = OMIT,
         raw_prompting: typing.Optional[bool] = OMIT,
@@ -210,6 +211,8 @@ def chat_stream(
             - p: typing.Optional[float]. Ensures that only the most likely tokens, with total probability mass of `p`, are considered for generation at each step. If both `k` and `p` are enabled, `p` acts after `k`.
                                          Defaults to `0.75`. min value of `0.01`, max value of `0.99`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinism cannot be totally guaranteed.
+
             - frequency_penalty: typing.Optional[float]. Defaults to `0.0`, min value of `0.0`, max value of `1.0`.
 
                                                          Used to reduce repetitiveness of generated tokens. The higher the value, the stronger a penalty is applied to previously present tokens, proportional to how many times they have already appeared in the prompt or prior generation.
@@ -353,6 +356,8 @@ def chat_stream(
             _request["k"] = k
         if p is not OMIT:
             _request["p"] = p
+        if seed is not OMIT:
+            _request["seed"] = seed
         if frequency_penalty is not OMIT:
             _request["frequency_penalty"] = frequency_penalty
         if presence_penalty is not OMIT:
@@ -420,6 +425,7 @@ def chat(
         max_tokens: typing.Optional[int] = OMIT,
         k: typing.Optional[int] = OMIT,
         p: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         frequency_penalty: typing.Optional[float] = OMIT,
         presence_penalty: typing.Optional[float] = OMIT,
         raw_prompting: typing.Optional[bool] = OMIT,
@@ -502,6 +508,8 @@ def chat(
             - p: typing.Optional[float]. Ensures that only the most likely tokens, with total probability mass of `p`, are considered for generation at each step. If both `k` and `p` are enabled, `p` acts after `k`.
                                          Defaults to `0.75`. min value of `0.01`, max value of `0.99`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinism cannot be totally guaranteed.
+
             - frequency_penalty: typing.Optional[float]. Defaults to `0.0`, min value of `0.0`, max value of `1.0`.
 
                                                          Used to reduce repetitiveness of generated tokens. The higher the value, the stronger a penalty is applied to previously present tokens, proportional to how many times they have already appeared in the prompt or prior generation.
@@ -588,6 +596,8 @@ def chat(
             _request["k"] = k
         if p is not OMIT:
             _request["p"] = p
+        if seed is not OMIT:
+            _request["seed"] = seed
         if frequency_penalty is not OMIT:
             _request["frequency_penalty"] = frequency_penalty
         if presence_penalty is not OMIT:
@@ -643,6 +653,7 @@ def generate_stream(
         max_tokens: typing.Optional[int] = OMIT,
         truncate: typing.Optional[GenerateStreamRequestTruncate] = OMIT,
         temperature: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         preset: typing.Optional[str] = OMIT,
         end_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
         stop_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
@@ -683,6 +694,8 @@ def generate_stream(
             - temperature: typing.Optional[float]. A non-negative float that tunes the degree of randomness in generation. Lower temperatures mean less random generations. See [Temperature](/temperature-wiki) for more details.
                                                    Defaults to `0.75`, min value of `0.0`, max value of `5.0`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinsim cannot be totally guaranteed.
+
             - preset: typing.Optional[str]. Identifier of a custom preset. A preset is a combination of parameters, such as prompt, temperature etc. You can create presets in the [playground](https://dashboard.cohere.ai/playground/generate).
                                             When a preset is specified, the `prompt` parameter becomes optional, and any included parameters will override the preset's parameters.
 
@@ -751,6 +764,8 @@ def generate_stream(
             _request["truncate"] = truncate
         if temperature is not OMIT:
             _request["temperature"] = temperature
+        if seed is not OMIT:
+            _request["seed"] = seed
         if preset is not OMIT:
             _request["preset"] = preset
         if end_sequences is not OMIT:
@@ -823,6 +838,7 @@ def generate(
         max_tokens: typing.Optional[int] = OMIT,
         truncate: typing.Optional[GenerateRequestTruncate] = OMIT,
         temperature: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         preset: typing.Optional[str] = OMIT,
         end_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
         stop_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
@@ -863,6 +879,8 @@ def generate(
             - temperature: typing.Optional[float]. A non-negative float that tunes the degree of randomness in generation. Lower temperatures mean less random generations. See [Temperature](/temperature-wiki) for more details.
                                                    Defaults to `0.75`, min value of `0.0`, max value of `5.0`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinsim cannot be totally guaranteed.
+
             - preset: typing.Optional[str]. Identifier of a custom preset. A preset is a combination of parameters, such as prompt, temperature etc. You can create presets in the [playground](https://dashboard.cohere.ai/playground/generate).
                                             When a preset is specified, the `prompt` parameter becomes optional, and any included parameters will override the preset's parameters.
 
@@ -917,6 +935,8 @@ def generate(
             _request["truncate"] = truncate
         if temperature is not OMIT:
             _request["temperature"] = temperature
+        if seed is not OMIT:
+            _request["seed"] = seed
         if preset is not OMIT:
             _request["preset"] = preset
         if end_sequences is not OMIT:
@@ -1608,6 +1628,7 @@ async def chat_stream(
         max_tokens: typing.Optional[int] = OMIT,
         k: typing.Optional[int] = OMIT,
         p: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         frequency_penalty: typing.Optional[float] = OMIT,
         presence_penalty: typing.Optional[float] = OMIT,
         raw_prompting: typing.Optional[bool] = OMIT,
@@ -1690,6 +1711,8 @@ async def chat_stream(
             - p: typing.Optional[float]. Ensures that only the most likely tokens, with total probability mass of `p`, are considered for generation at each step. If both `k` and `p` are enabled, `p` acts after `k`.
                                          Defaults to `0.75`. min value of `0.01`, max value of `0.99`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinism cannot be totally guaranteed.
+
             - frequency_penalty: typing.Optional[float]. Defaults to `0.0`, min value of `0.0`, max value of `1.0`.
 
                                                          Used to reduce repetitiveness of generated tokens. The higher the value, the stronger a penalty is applied to previously present tokens, proportional to how many times they have already appeared in the prompt or prior generation.
@@ -1833,6 +1856,8 @@ async def chat_stream(
             _request["k"] = k
         if p is not OMIT:
             _request["p"] = p
+        if seed is not OMIT:
+            _request["seed"] = seed
         if frequency_penalty is not OMIT:
             _request["frequency_penalty"] = frequency_penalty
         if presence_penalty is not OMIT:
@@ -1900,6 +1925,7 @@ async def chat(
         max_tokens: typing.Optional[int] = OMIT,
         k: typing.Optional[int] = OMIT,
         p: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         frequency_penalty: typing.Optional[float] = OMIT,
         presence_penalty: typing.Optional[float] = OMIT,
         raw_prompting: typing.Optional[bool] = OMIT,
@@ -1982,6 +2008,8 @@ async def chat(
             - p: typing.Optional[float]. Ensures that only the most likely tokens, with total probability mass of `p`, are considered for generation at each step. If both `k` and `p` are enabled, `p` acts after `k`.
                                          Defaults to `0.75`. min value of `0.01`, max value of `0.99`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinism cannot be totally guaranteed.
+
             - frequency_penalty: typing.Optional[float]. Defaults to `0.0`, min value of `0.0`, max value of `1.0`.
 
                                                          Used to reduce repetitiveness of generated tokens. The higher the value, the stronger a penalty is applied to previously present tokens, proportional to how many times they have already appeared in the prompt or prior generation.
@@ -2068,6 +2096,8 @@ async def chat(
             _request["k"] = k
         if p is not OMIT:
             _request["p"] = p
+        if seed is not OMIT:
+            _request["seed"] = seed
         if frequency_penalty is not OMIT:
             _request["frequency_penalty"] = frequency_penalty
         if presence_penalty is not OMIT:
@@ -2123,6 +2153,7 @@ async def generate_stream(
         max_tokens: typing.Optional[int] = OMIT,
         truncate: typing.Optional[GenerateStreamRequestTruncate] = OMIT,
         temperature: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         preset: typing.Optional[str] = OMIT,
         end_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
         stop_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
@@ -2163,6 +2194,8 @@ async def generate_stream(
             - temperature: typing.Optional[float]. A non-negative float that tunes the degree of randomness in generation. Lower temperatures mean less random generations. See [Temperature](/temperature-wiki) for more details.
                                                    Defaults to `0.75`, min value of `0.0`, max value of `5.0`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinsim cannot be totally guaranteed.
+
             - preset: typing.Optional[str]. Identifier of a custom preset. A preset is a combination of parameters, such as prompt, temperature etc. You can create presets in the [playground](https://dashboard.cohere.ai/playground/generate).
                                             When a preset is specified, the `prompt` parameter becomes optional, and any included parameters will override the preset's parameters.
 
@@ -2231,6 +2264,8 @@ async def generate_stream(
             _request["truncate"] = truncate
         if temperature is not OMIT:
             _request["temperature"] = temperature
+        if seed is not OMIT:
+            _request["seed"] = seed
         if preset is not OMIT:
             _request["preset"] = preset
         if end_sequences is not OMIT:
@@ -2303,6 +2338,7 @@ async def generate(
         max_tokens: typing.Optional[int] = OMIT,
         truncate: typing.Optional[GenerateRequestTruncate] = OMIT,
         temperature: typing.Optional[float] = OMIT,
+        seed: typing.Optional[float] = OMIT,
         preset: typing.Optional[str] = OMIT,
         end_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
         stop_sequences: typing.Optional[typing.Sequence[str]] = OMIT,
@@ -2343,6 +2379,8 @@ async def generate(
             - temperature: typing.Optional[float]. A non-negative float that tunes the degree of randomness in generation. Lower temperatures mean less random generations. See [Temperature](/temperature-wiki) for more details.
                                                    Defaults to `0.75`, min value of `0.0`, max value of `5.0`.
 
+            - seed: typing.Optional[float]. If specified, the backend will make a best effort to sample tokens deterministically, such that repeated requests with the same seed and parameters should return the same result. However, determinsim cannot be totally guaranteed.
+
             - preset: typing.Optional[str]. Identifier of a custom preset. A preset is a combination of parameters, such as prompt, temperature etc. You can create presets in the [playground](https://dashboard.cohere.ai/playground/generate).
                                             When a preset is specified, the `prompt` parameter becomes optional, and any included parameters will override the preset's parameters.
 
@@ -2397,6 +2435,8 @@ async def generate(
             _request["truncate"] = truncate
         if temperature is not OMIT:
             _request["temperature"] = temperature
+        if seed is not OMIT:
+            _request["seed"] = seed
         if preset is not OMIT:
             _request["preset"] = preset
         if end_sequences is not OMIT:

src/cohere/core/client_wrapper.py

@@ -23,7 +23,7 @@ def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "cohere",
-            "X-Fern-SDK-Version": "5.0.0a11",
+            "X-Fern-SDK-Version": "5.0.0a12",
         }
         if self._client_name is not None:
             headers["X-Client-Name"] = self._client_name

src/cohere/embed_jobs/types/create_embed_job_request_truncate.py

@@ -2,4 +2,4 @@
 
 import typing
 
-CreateEmbedJobRequestTruncate = typing.Literal["START", "END"]
+CreateEmbedJobRequestTruncate = typing.Union[typing.AnyStr, typing.Literal["START", "END"]]

src/cohere/types/auth_token_type.py

@@ -2,4 +2,4 @@
 
 import typing
 
-AuthTokenType = typing.Literal["bearer", "basic", "noscheme"]
+AuthTokenType = typing.Union[typing.AnyStr, typing.Literal["bearer", "basic", "noscheme"]]

src/cohere/types/chat_message_role.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ChatMessageRole = typing.Literal["CHATBOT", "SYSTEM", "USER"]
+ChatMessageRole = typing.Union[typing.AnyStr, typing.Literal["CHATBOT", "SYSTEM", "USER"]]

src/cohere/types/chat_request_citation_quality.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ChatRequestCitationQuality = typing.Literal["fast", "accurate"]
+ChatRequestCitationQuality = typing.Union[typing.AnyStr, typing.Literal["fast", "accurate"]]

src/cohere/types/chat_request_prompt_truncation.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ChatRequestPromptTruncation = typing.Literal["OFF", "AUTO", "AUTO_PRESERVE_ORDER"]
+ChatRequestPromptTruncation = typing.Union[typing.AnyStr, typing.Literal["OFF", "AUTO", "AUTO_PRESERVE_ORDER"]]

src/cohere/types/chat_stream_end_event_finish_reason.py

@@ -2,4 +2,6 @@
 
 import typing
 
-ChatStreamEndEventFinishReason = typing.Literal["COMPLETE", "ERROR_LIMIT", "MAX_TOKENS", "ERROR", "ERROR_TOXIC"]
+ChatStreamEndEventFinishReason = typing.Union[
+    typing.AnyStr, typing.Literal["COMPLETE", "ERROR_LIMIT", "MAX_TOKENS", "ERROR", "ERROR_TOXIC"]
+]

src/cohere/types/chat_stream_request_citation_quality.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ChatStreamRequestCitationQuality = typing.Literal["fast", "accurate"]
+ChatStreamRequestCitationQuality = typing.Union[typing.AnyStr, typing.Literal["fast", "accurate"]]

src/cohere/types/chat_stream_request_prompt_truncation.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ChatStreamRequestPromptTruncation = typing.Literal["OFF", "AUTO", "AUTO_PRESERVE_ORDER"]
+ChatStreamRequestPromptTruncation = typing.Union[typing.AnyStr, typing.Literal["OFF", "AUTO", "AUTO_PRESERVE_ORDER"]]

src/cohere/types/classify_request_truncate.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ClassifyRequestTruncate = typing.Literal["NONE", "START", "END"]
+ClassifyRequestTruncate = typing.Union[typing.AnyStr, typing.Literal["NONE", "START", "END"]]

src/cohere/types/classify_response_classifications_item_classification_type.py

@@ -2,4 +2,6 @@
 
 import typing
 
-ClassifyResponseClassificationsItemClassificationType = typing.Literal["single-label", "multi-label"]
+ClassifyResponseClassificationsItemClassificationType = typing.Union[
+    typing.AnyStr, typing.Literal["single-label", "multi-label"]
+]

src/cohere/types/compatible_endpoint.py

@@ -2,4 +2,6 @@
 
 import typing
 
-CompatibleEndpoint = typing.Literal["chat", "embed", "classify", "summarize", "rerank", "rate", "generate"]
+CompatibleEndpoint = typing.Union[
+    typing.AnyStr, typing.Literal["chat", "embed", "classify", "summarize", "rerank", "rate", "generate"]
+]

src/cohere/types/connector_auth_status.py

@@ -2,4 +2,4 @@
 
 import typing
 
-ConnectorAuthStatus = typing.Literal["valid", "expired"]
+ConnectorAuthStatus = typing.Union[typing.AnyStr, typing.Literal["valid", "expired"]]

src/cohere/types/dataset_type.py

@@ -2,14 +2,17 @@
 
 import typing
 
-DatasetType = typing.Literal[
-    "embed-input",
-    "embed-result",
-    "cluster-result",
-    "cluster-outliers",
-    "reranker-finetune-input",
-    "prompt-completion-finetune-input",
-    "single-label-classification-finetune-input",
-    "chat-finetune-input",
-    "multi-label-classification-finetune-input",
+DatasetType = typing.Union[
+    typing.AnyStr,
+    typing.Literal[
+        "embed-input",
+        "embed-result",
+        "cluster-result",
+        "cluster-outliers",
+        "reranker-finetune-input",
+        "prompt-completion-finetune-input",
+        "single-label-classification-finetune-input",
+        "chat-finetune-input",
+        "multi-label-classification-finetune-input",
+    ],
 ]

src/cohere/types/dataset_validation_status.py

@@ -2,4 +2,6 @@
 
 import typing
 
-DatasetValidationStatus = typing.Literal["unknown", "queued", "processing", "failed", "validated", "skipped"]
+DatasetValidationStatus = typing.Union[
+    typing.AnyStr, typing.Literal["unknown", "queued", "processing", "failed", "validated", "skipped"]
+]

src/cohere/types/embed_input_type.py

@@ -2,4 +2,6 @@
 
 import typing
 
-EmbedInputType = typing.Literal["search_document", "search_query", "classification", "clustering"]
+EmbedInputType = typing.Union[
+    typing.AnyStr, typing.Literal["search_document", "search_query", "classification", "clustering"]
+]

src/cohere/types/embed_job_status.py

@@ -2,4 +2,6 @@
 
 import typing
 
-EmbedJobStatus = typing.Literal["processing", "complete", "cancelling", "cancelled", "failed"]
+EmbedJobStatus = typing.Union[
+    typing.AnyStr, typing.Literal["processing", "complete", "cancelling", "cancelled", "failed"]
+]