together 1.5.34__py3-none-any.whl → 2.0.0a6__py3-none-any.whl

together/types/code_interpreter_execute_params.py

@@ -0,0 +1,45 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Iterable
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["CodeInterpreterExecuteParams", "File"]
+
+
+class CodeInterpreterExecuteParams(TypedDict, total=False):
+    code: Required[str]
+    """Code snippet to execute."""
+
+    language: Required[Literal["python"]]
+    """Programming language for the code to execute.
+
+    Currently only supports Python, but more will be added.
+    """
+
+    files: Iterable[File]
+    """Files to upload to the session.
+
+    If present, files will be uploaded before executing the given code.
+    """
+
+    session_id: str
+    """Identifier of the current session.
+
+    Used to make follow-up calls. Requests will return an error if the session does
+    not belong to the caller or has expired.
+    """
+
+
+class File(TypedDict, total=False):
+    content: Required[str]
+
+    encoding: Required[Literal["string", "base64"]]
+    """Encoding of the file content.
+
+    Use `string` for text files such as code, and `base64` for binary files, such as
+    images.
+    """
+
+    name: Required[str]

together/types/completion.py

@@ -0,0 +1,42 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+from .._models import BaseModel
+from .log_probs import LogProbs
+from .chat.chat_completion_usage import ChatCompletionUsage
+
+__all__ = ["Completion", "Choice", "Prompt"]
+
+
+class Choice(BaseModel):
+    finish_reason: Optional[Literal["stop", "eos", "length", "tool_calls", "function_call"]] = None
+
+    logprobs: Optional[LogProbs] = None
+
+    seed: Optional[int] = None
+
+    text: Optional[str] = None
+
+
+class Prompt(BaseModel):
+    logprobs: Optional[LogProbs] = None
+
+    text: Optional[str] = None
+
+
+class Completion(BaseModel):
+    id: str
+
+    choices: List[Choice]
+
+    created: int
+
+    model: str
+
+    object: Literal["text.completion"]
+
+    usage: Optional[ChatCompletionUsage] = None
+
+    prompt: Optional[List[Prompt]] = None

together/types/completion_chunk.py

@@ -0,0 +1,66 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+from .._models import BaseModel
+from .tool_choice import ToolChoice
+from .chat.chat_completion_usage import ChatCompletionUsage
+
+__all__ = ["CompletionChunk", "Token", "Choice", "ChoiceDelta", "ChoiceDeltaFunctionCall"]
+
+
+class Token(BaseModel):
+    id: int
+
+    logprob: float
+
+    special: bool
+
+    text: str
+
+
+class ChoiceDeltaFunctionCall(BaseModel):
+    arguments: str
+
+    name: str
+
+
+class ChoiceDelta(BaseModel):
+    role: Literal["system", "user", "assistant", "function", "tool"]
+
+    content: Optional[str] = None
+
+    function_call: Optional[ChoiceDeltaFunctionCall] = None
+
+    reasoning: Optional[str] = None
+
+    token_id: Optional[int] = None
+
+    tool_calls: Optional[List[ToolChoice]] = None
+
+
+class Choice(BaseModel):
+    index: int
+
+    delta: Optional[ChoiceDelta] = None
+
+    text: Optional[str] = None
+
+
+class CompletionChunk(BaseModel):
+    id: str
+
+    token: Token
+
+    choices: List[Choice]
+
+    finish_reason: Optional[Literal["stop", "eos", "length", "tool_calls", "function_call"]] = None
+
+    usage: Optional[ChatCompletionUsage] = None
+
+    created: Optional[int] = None
+
+    object: Optional[Literal["completion.chunk"]] = None
+
+    seed: Optional[int] = None

together/types/completion_create_params.py

@@ -0,0 +1,138 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Dict, Union
+from typing_extensions import Literal, Required, TypedDict
+
+from .._types import SequenceNotStr
+
+__all__ = ["CompletionCreateParamsBase", "CompletionCreateParamsNonStreaming", "CompletionCreateParamsStreaming"]
+
+
+class CompletionCreateParamsBase(TypedDict, total=False):
+    model: Required[
+        Union[
+            Literal[
+                "meta-llama/Llama-2-70b-hf",
+                "mistralai/Mistral-7B-v0.1",
+                "mistralai/Mixtral-8x7B-v0.1",
+                "Meta-Llama/Llama-Guard-7b",
+            ],
+            str,
+        ]
+    ]
+    """The name of the model to query.
+
+    [See all of Together AI's chat models](https://docs.together.ai/docs/serverless-models#chat-models)
+    """
+
+    prompt: Required[str]
+    """A string providing context for the model to complete."""
+
+    echo: bool
+    """If true, the response will contain the prompt.
+
+    Can be used with `logprobs` to return prompt logprobs.
+    """
+
+    frequency_penalty: float
+    """
+    A number between -2.0 and 2.0 where a positive value decreases the likelihood of
+    repeating tokens that have already been mentioned.
+    """
+
+    logit_bias: Dict[str, float]
+    """Adjusts the likelihood of specific tokens appearing in the generated output."""
+
+    logprobs: int
+    """
+    An integer between 0 and 20 of the top k tokens to return log probabilities for
+    at each generation step, instead of just the sampled token. Log probabilities
+    help assess model confidence in token predictions.
+    """
+
+    max_tokens: int
+    """The maximum number of tokens to generate."""
+
+    min_p: float
+    """A number between 0 and 1 that can be used as an alternative to top-p and top-k."""
+
+    n: int
+    """The number of completions to generate for each prompt."""
+
+    presence_penalty: float
+    """
+    A number between -2.0 and 2.0 where a positive value increases the likelihood of
+    a model talking about new topics.
+    """
+
+    repetition_penalty: float
+    """
+    A number that controls the diversity of generated text by reducing the
+    likelihood of repeated sequences. Higher values decrease repetition.
+    """
+
+    safety_model: Union[Literal["Meta-Llama/Llama-Guard-7b"], str]
+    """The name of the moderation model used to validate tokens.
+
+    Choose from the available moderation models found
+    [here](https://docs.together.ai/docs/inference-models#moderation-models).
+    """
+
+    seed: int
+    """Seed value for reproducibility."""
+
+    stop: SequenceNotStr[str]
+    """A list of string sequences that will truncate (stop) inference text output.
+
+    For example, "</s>" will stop generation as soon as the model generates the
+    given token.
+    """
+
+    temperature: float
+    """
+    A decimal number from 0-1 that determines the degree of randomness in the
+    response. A temperature less than 1 favors more correctness and is appropriate
+    for question answering or summarization. A value closer to 1 introduces more
+    randomness in the output.
+    """
+
+    top_k: int
+    """
+    An integer that's used to limit the number of choices for the next predicted
+    word or token. It specifies the maximum number of tokens to consider at each
+    step, based on their probability of occurrence. This technique helps to speed up
+    the generation process and can improve the quality of the generated text by
+    focusing on the most likely options.
+    """
+
+    top_p: float
+    """
+    A percentage (also called the nucleus parameter) that's used to dynamically
+    adjust the number of choices for each predicted token based on the cumulative
+    probabilities. It specifies a probability threshold below which all less likely
+    tokens are filtered out. This technique helps maintain diversity and generate
+    more fluent and natural-sounding text.
+    """
+
+
+class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase, total=False):
+    stream: Literal[False]
+    """
+    If true, stream tokens as Server-Sent Events as the model generates them instead
+    of waiting for the full model response. The stream terminates with
+    `data: [DONE]`. If false, return a single JSON object containing the results.
+    """
+
+
+class CompletionCreateParamsStreaming(CompletionCreateParamsBase):
+    stream: Required[Literal[True]]
+    """
+    If true, stream tokens as Server-Sent Events as the model generates them instead
+    of waiting for the full model response. The stream terminates with
+    `data: [DONE]`. If false, return a single JSON object containing the results.
+    """
+
+
+CompletionCreateParams = Union[CompletionCreateParamsNonStreaming, CompletionCreateParamsStreaming]

together/types/dedicated_endpoint.py

@@ -0,0 +1,44 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from datetime import datetime
+from typing_extensions import Literal
+
+from .._models import BaseModel
+from .autoscaling import Autoscaling
+
+__all__ = ["DedicatedEndpoint"]
+
+
+class DedicatedEndpoint(BaseModel):
+    id: str
+    """Unique identifier for the endpoint"""
+
+    autoscaling: Autoscaling
+    """Configuration for automatic scaling of the endpoint"""
+
+    created_at: datetime
+    """Timestamp when the endpoint was created"""
+
+    display_name: str
+    """Human-readable name for the endpoint"""
+
+    hardware: str
+    """The hardware configuration used for this endpoint"""
+
+    model: str
+    """The model deployed on this endpoint"""
+
+    name: str
+    """System name for the endpoint"""
+
+    object: Literal["endpoint"]
+    """The type of object"""
+
+    owner: str
+    """The owner of this endpoint"""
+
+    state: Literal["PENDING", "STARTING", "STARTED", "STOPPING", "STOPPED", "ERROR"]
+    """Current state of the endpoint"""
+
+    type: Literal["dedicated"]
+    """The type of endpoint"""

together/types/embedding.py

@@ -0,0 +1,24 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List
+from typing_extensions import Literal
+
+from .._models import BaseModel
+
+__all__ = ["Embedding", "Data"]
+
+
+class Data(BaseModel):
+    embedding: List[float]
+
+    index: int
+
+    object: Literal["embedding"]
+
+
+class Embedding(BaseModel):
+    data: List[Data]
+
+    model: str
+
+    object: Literal["list"]

together/types/embedding_create_params.py

@@ -0,0 +1,31 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import Literal, Required, TypedDict
+
+from .._types import SequenceNotStr
+
+__all__ = ["EmbeddingCreateParams"]
+
+
+class EmbeddingCreateParams(TypedDict, total=False):
+    input: Required[Union[str, SequenceNotStr[str]]]
+    """A string providing the text for the model to embed."""
+
+    model: Required[
+        Union[
+            Literal[
+                "WhereIsAI/UAE-Large-V1",
+                "BAAI/bge-large-en-v1.5",
+                "BAAI/bge-base-en-v1.5",
+                "togethercomputer/m2-bert-80M-8k-retrieval",
+            ],
+            str,
+        ]
+    ]
+    """The name of the embedding model to use.
+
+    [See all of Together AI's embedding models](https://docs.together.ai/docs/serverless-models#embedding-models)
+    """

together/types/endpoint_create_params.py

@@ -0,0 +1,43 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Optional
+from typing_extensions import Literal, Required, TypedDict
+
+from .autoscaling_param import AutoscalingParam
+
+__all__ = ["EndpointCreateParams"]
+
+
+class EndpointCreateParams(TypedDict, total=False):
+    autoscaling: Required[AutoscalingParam]
+    """Configuration for automatic scaling of the endpoint"""
+
+    hardware: Required[str]
+    """The hardware configuration to use for this endpoint"""
+
+    model: Required[str]
+    """The model to deploy on this endpoint"""
+
+    availability_zone: str
+    """Create the endpoint in a specified availability zone (e.g., us-central-4b)"""
+
+    disable_prompt_cache: bool
+    """Whether to disable the prompt cache for this endpoint"""
+
+    disable_speculative_decoding: bool
+    """Whether to disable speculative decoding for this endpoint"""
+
+    display_name: str
+    """A human-readable name for the endpoint"""
+
+    inactive_timeout: Optional[int]
+    """
+    The number of minutes of inactivity after which the endpoint will be
+    automatically stopped. Set to null, omit or set to 0 to disable automatic
+    timeout.
+    """
+
+    state: Literal["STARTED", "STOPPED"]
+    """The desired state of the endpoint"""

together/types/endpoint_list_avzones_response.py

@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List
+
+from .._models import BaseModel
+
+__all__ = ["EndpointListAvzonesResponse"]
+
+
+class EndpointListAvzonesResponse(BaseModel):
+    avzones: List[str]

together/types/endpoint_list_params.py

@@ -0,0 +1,18 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import Literal, TypedDict
+
+__all__ = ["EndpointListParams"]
+
+
+class EndpointListParams(TypedDict, total=False):
+    mine: bool
+    """If true, return only endpoints owned by the caller"""
+
+    type: Literal["dedicated", "serverless"]
+    """Filter endpoints by type"""
+
+    usage_type: Literal["on-demand", "reserved"]
+    """Filter endpoints by usage type"""

together/types/endpoint_list_response.py

@@ -0,0 +1,41 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List
+from datetime import datetime
+from typing_extensions import Literal
+
+from .._models import BaseModel
+
+__all__ = ["EndpointListResponse", "Data"]
+
+
+class Data(BaseModel):
+    id: str
+    """Unique identifier for the endpoint"""
+
+    created_at: datetime
+    """Timestamp when the endpoint was created"""
+
+    model: str
+    """The model deployed on this endpoint"""
+
+    name: str
+    """System name for the endpoint"""
+
+    object: Literal["endpoint"]
+    """The type of object"""
+
+    owner: str
+    """The owner of this endpoint"""
+
+    state: Literal["PENDING", "STARTING", "STARTED", "STOPPING", "STOPPED", "ERROR"]
+    """Current state of the endpoint"""
+
+    type: Literal["serverless", "dedicated"]
+    """The type of endpoint"""
+
+
+class EndpointListResponse(BaseModel):
+    data: List[Data]
+
+    object: Literal["list"]

together/types/endpoint_update_params.py

@@ -0,0 +1,27 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Optional
+from typing_extensions import Literal, TypedDict
+
+from .autoscaling_param import AutoscalingParam
+
+__all__ = ["EndpointUpdateParams"]
+
+
+class EndpointUpdateParams(TypedDict, total=False):
+    autoscaling: AutoscalingParam
+    """New autoscaling configuration for the endpoint"""
+
+    display_name: str
+    """A human-readable name for the endpoint"""
+
+    inactive_timeout: Optional[int]
+    """
+    The number of minutes of inactivity after which the endpoint will be
+    automatically stopped. Set to 0 to disable automatic timeout.
+    """
+
+    state: Literal["STARTED", "STOPPED"]
+    """The desired state of the endpoint"""