openai.calls¶
A module for calling OpenAI's Chat Completion models.
BaseCall
¶
Bases: BasePrompt
, Generic[BaseCallResponseT, BaseCallResponseChunkT, BaseToolT, MessageParamT]
, ABC
The base class abstract interface for calling LLMs.
Source code in mirascope/base/calls.py
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 |
|
call(retries=0, **kwargs)
abstractmethod
¶
A call to an LLM.
An implementation of this function must return a response that extends
BaseCallResponse
. This ensures a consistent API and convenience across e.g.
different model providers.
Source code in mirascope/base/calls.py
call_async(retries=0, **kwargs)
abstractmethod
async
¶
An asynchronous call to an LLM.
An implementation of this function must return a response that extends
BaseCallResponse
. This ensures a consistent API and convenience across e.g.
different model providers.
Source code in mirascope/base/calls.py
from_prompt(prompt_type, call_params)
classmethod
¶
Returns a call_type generated dynamically from this base call.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
prompt_type |
type[BasePromptT]
|
The prompt class to use for the call. Properties and class variables of this class will be used to create the new call class. Must be a class that can be instantiated. |
required |
call_params |
BaseCallParams
|
The call params to use for the call. |
required |
Returns:
Type | Description |
---|---|
type[BasePromptT]
|
A new call class with new call_type. |
Source code in mirascope/base/calls.py
stream(retries=0, **kwargs)
abstractmethod
¶
A call to an LLM that streams the response in chunks.
An implementation of this function must yield response chunks that extend
BaseCallResponseChunk
. This ensures a consistent API and convenience across
e.g. different model providers.
Source code in mirascope/base/calls.py
stream_async(retries=0, **kwargs)
abstractmethod
async
¶
A asynchronous call to an LLM that streams the response in chunks.
An implementation of this function must yield response chunks that extend
BaseCallResponseChunk
. This ensures a consistent API and convenience across
e.g. different model providers.
Source code in mirascope/base/calls.py
MessageRole
¶
Bases: _Enum
Roles that the BasePrompt
messages parser can parse from the template.
SYSTEM: A system message. USER: A user message. ASSISTANT: A message response from the assistant or chat client. MODEL: A message response from the assistant or chat client. Model is used by Google's Gemini instead of assistant, which doesn't have system messages. CHATBOT: A message response from the chat client. Chatbot is used by Cohere instead of assistant. TOOL: A message representing the output of calling a tool.
Source code in mirascope/enums.py
OpenAICall
¶
Bases: BaseCall[OpenAICallResponse, OpenAICallResponseChunk, OpenAITool, ChatCompletionUserMessageParam]
A base class for calling OpenAI's Chat Completion models.
Example:
from mirascope.openai import OpenAICall
class BookRecommender(OpenAICall):
prompt_template = "Please recommend a {genre} book"
genre: str
response = BookRecommender(genre="fantasy").call()
print(response.content)
#> There are many great books to read, it ultimately depends...
Source code in mirascope/openai/calls.py
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 |
|
call(retries=0, **kwargs)
¶
Makes a call to the model using this OpenAICall
instance.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
retries |
Union[int, Retrying]
|
An integer for the number of times to retry the call or
a |
0
|
**kwargs |
Any
|
Additional keyword arguments parameters to pass to the call. These
will override any existing arguments in |
{}
|
Returns:
Type | Description |
---|---|
OpenAICallResponse
|
A |
Raises:
Type | Description |
---|---|
OpenAIError
|
raises any OpenAI errors, see: https://platform.openai.com/docs/guides/error-codes/api-errors |
Source code in mirascope/openai/calls.py
call_async(retries=0, **kwargs)
async
¶
Makes an asynchronous call to the model using this OpenAICall
.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
retries |
Union[int, AsyncRetrying]
|
An integer for the number of times to retry the call or
a |
0
|
**kwargs |
Any
|
Additional keyword arguments parameters to pass to the call. These
will override any existing arguments in |
{}
|
Returns:
Type | Description |
---|---|
OpenAICallResponse
|
An |
Raises:
Type | Description |
---|---|
OpenAIError
|
raises any OpenAI errors, see: https://platform.openai.com/docs/guides/error-codes/api-errors |
Source code in mirascope/openai/calls.py
messages()
¶
Returns the template as a formatted list of messages.
Source code in mirascope/openai/calls.py
stream(retries=0, **kwargs)
¶
Streams the response for a call using this OpenAICall
.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
retries |
Union[int, Retrying]
|
An integer for the number of times to retry the call or
a |
0
|
**kwargs |
Any
|
Additional keyword arguments parameters to pass to the call. These
will override any existing arguments in |
{}
|
Yields:
Type | Description |
---|---|
OpenAICallResponseChunk
|
A |
Raises:
Type | Description |
---|---|
OpenAIError
|
raises any OpenAI errors, see: https://platform.openai.com/docs/guides/error-codes/api-errors |
Source code in mirascope/openai/calls.py
stream_async(retries=0, **kwargs)
async
¶
Streams the response for an asynchronous call using this OpenAICall
.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
retries |
Union[int, AsyncRetrying]
|
An integer for the number of times to retry the call or
a |
0
|
**kwargs |
Any
|
Additional keyword arguments parameters to pass to the call. These
will override any existing arguments in |
{}
|
Yields:
Type | Description |
---|---|
AsyncGenerator[OpenAICallResponseChunk, None]
|
A |
Raises:
Type | Description |
---|---|
OpenAIError
|
raises any OpenAI errors, see: https://platform.openai.com/docs/guides/error-codes/api-errors |
Source code in mirascope/openai/calls.py
OpenAICallParams
¶
Bases: BaseCallParams[OpenAITool]
The parameters to use when calling the OpenAI API.
Source code in mirascope/openai/types.py
kwargs(tool_type=OpenAITool, exclude=None)
¶
Returns the keyword argument call parameters.
OpenAICallResponse
¶
Bases: BaseCallResponse[ChatCompletion, OpenAITool]
A convenience wrapper around the OpenAI ChatCompletion
response.
When using Mirascope's convenience wrappers to interact with OpenAI models via
OpenAICall
, responses using OpenAICall.call()
will return a
OpenAICallResponse
, whereby the implemented properties allow for simpler syntax
and a convenient developer experience.
Example:
from mirascope.openai import OpenAICall
class BookRecommender(OpenAICall):
prompt_template = "Please recommend a {genre} book"
genre: str
response = Bookrecommender(genre="fantasy").call()
print(response.content)
#> The Name of the Wind
print(response.message)
#> ChatCompletionMessage(content='The Name of the Wind', role='assistant',
# function_call=None, tool_calls=None)
print(response.choices)
#> [Choice(finish_reason='stop', index=0, logprobs=None,
# message=ChatCompletionMessage(content='The Name of the Wind', role='assistant',
# function_call=None, tool_calls=None))]
Source code in mirascope/openai/types.py
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 |
|
choice: Choice
property
¶
Returns the 0th choice.
choices: list[Choice]
property
¶
Returns the array of chat completion choices.
content: str
property
¶
Returns the content of the chat completion for the 0th choice.
finish_reasons: list[str]
property
¶
Returns the finish reasons of the response.
id: str
property
¶
Returns the id of the response.
input_tokens: Optional[int]
property
¶
Returns the number of input tokens.
message: ChatCompletionMessage
property
¶
Returns the message of the chat completion for the 0th choice.
message_param: ChatCompletionAssistantMessageParam
property
¶
Returns the assistants's response as a message parameter.
model: str
property
¶
Returns the name of the response model.
output_tokens: Optional[int]
property
¶
Returns the number of output tokens.
tool: Optional[OpenAITool]
property
¶
Returns the 0th tool for the 0th choice message.
Raises:
Type | Description |
---|---|
ValidationError
|
if the tool call doesn't match the tool's schema. |
tool_calls: Optional[list[ChatCompletionMessageToolCall]]
property
¶
Returns the tool calls for the 0th choice message.
tools: Optional[list[OpenAITool]]
property
¶
Returns the tools for the 0th choice message.
Raises:
Type | Description |
---|---|
ValidationError
|
if a tool call doesn't match the tool's schema. |
usage: Optional[CompletionUsage]
property
¶
Returns the usage of the chat completion.
dump()
¶
Dumps the response to a dictionary.
OpenAICallResponseChunk
¶
Bases: BaseCallResponseChunk[ChatCompletionChunk, OpenAITool]
Convenience wrapper around chat completion streaming chunks.
When using Mirascope's convenience wrappers to interact with OpenAI models via
OpenAICall.stream
, responses will return an OpenAICallResponseChunk
, whereby
the implemented properties allow for simpler syntax and a convenient developer
experience.
Example:
from mirascope.openai import OpenAICall
class Math(OpenAICall):
prompt_template = "What is 1 + 2?"
content = ""
for chunk in Math().stream():
content += chunk.content
print(content)
#> 1
# 1 +
# 1 + 2
# 1 + 2 equals
# 1 + 2 equals
# 1 + 2 equals 3
# 1 + 2 equals 3.
Source code in mirascope/openai/types.py
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 |
|
choice: ChunkChoice
property
¶
Returns the 0th choice.
choices: list[ChunkChoice]
property
¶
Returns the array of chat completion choices.
content: str
property
¶
Returns the content for the 0th choice delta.
delta: Optional[ChoiceDelta]
property
¶
Returns the delta for the 0th choice.
finish_reasons: list[str]
property
¶
Returns the finish reasons of the response.
id: str
property
¶
Returns the id of the response.
input_tokens: Optional[int]
property
¶
Returns the number of input tokens.
model: str
property
¶
Returns the name of the response model.
output_tokens: Optional[int]
property
¶
Returns the number of output tokens.
tool_calls: Optional[list[ChoiceDeltaToolCall]]
property
¶
Returns the partial tool calls for the 0th choice message.
The first list[ChoiceDeltaToolCall]
will contain the name of the tool and
index, and subsequent list[ChoiceDeltaToolCall]
s will contain the arguments
which will be strings that need to be concatenated with future
list[ChoiceDeltaToolCall]
s to form a complete JSON tool calls. The last
list[ChoiceDeltaToolCall]
will be None indicating end of stream.
usage: Optional[CompletionUsage]
property
¶
Returns the usage of the chat completion.
OpenAITool
¶
Bases: BaseTool[ChatCompletionMessageToolCall]
A base class for easy use of tools with the OpenAI Chat client.
OpenAITool
internally handles the logic that allows you to use tools with simple
calls such as OpenAICallResponse.tool
or OpenAITool.fn
, as seen in the
examples below.
Example:
from mirascope.openai import OpenAICall, OpenAICallParams
def animal_matcher(fav_food: str, fav_color: str) -> str:
"""Tells you your most likely favorite animal from personality traits.
Args:
fav_food: your favorite food.
fav_color: your favorite color.
Returns:
The animal most likely to be your favorite based on traits.
"""
return "Your favorite animal is the best one, a frog."
class AnimalMatcher(OpenAICall):
prompt_template = """
Tell me my favorite animal if my favorite food is {food} and my
favorite color is {color}.
"""
food: str
color: str
call_params = OpenAICallParams(tools=[animal_matcher])
response = AnimalMatcher(food="pizza", color="red").call
tool = response.tool
print(tool.fn(**tool.args))
#> Your favorite animal is the best one, a frog.
Source code in mirascope/openai/tools.py
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
|
from_base_type(base_type)
classmethod
¶
Constructs a OpenAITool
type from a BaseType
type.
from_fn(fn)
classmethod
¶
from_model(model)
classmethod
¶
Constructs a OpenAITool
type from a BaseModel
type.
from_tool_call(tool_call, allow_partial=False)
classmethod
¶
Extracts an instance of the tool constructed from a tool call response.
Given ChatCompletionMessageToolCall
from an OpenAI chat completion response,
takes its function arguments and creates an OpenAITool
instance from it.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
tool_call |
ChatCompletionMessageToolCall
|
The |
required |
allow_partial |
bool
|
Whether to allow partial JSON schemas. |
False
|
Returns:
Type | Description |
---|---|
OpenAITool
|
An instance of the tool constructed from the tool call. |
Raises:
Type | Description |
---|---|
ValidationError
|
if the tool call doesn't match the tool schema. |
Source code in mirascope/openai/tools.py
tool_schema()
classmethod
¶
Constructs a tool schema for use with the OpenAI Chat client.
A Mirascope OpenAITool
is deconstructed into a JSON schema, and relevant keys
are renamed to match the OpenAI ChatCompletionToolParam
schema used to make
function/tool calls in OpenAI API.
Returns:
Type | Description |
---|---|
ChatCompletionToolParam
|
The constructed |
Source code in mirascope/openai/tools.py
get_wrapped_async_client(client, self)
¶
Get a wrapped async client.
Source code in mirascope/base/ops_utils.py
get_wrapped_call(call, self, **kwargs)
¶
Wrap a call to add the llm_ops
parameter if it exists.
Source code in mirascope/base/ops_utils.py
get_wrapped_client(client, self)
¶
Get a wrapped client.
Source code in mirascope/base/ops_utils.py
openai_api_calculate_cost(usage, model='gpt-3.5-turbo-16k')
¶
Calculate the cost of a completion using the OpenAI API.
https://openai.com/pricing
Model Input Output gpt-4o $5.00 / 1M tokens $15.00 / 1M tokens gpt-4o-2024-05-13 $5.00 / 1M tokens $15.00 / 1M tokens gpt-4-turbo $10.00 / 1M tokens $30.00 / 1M tokens gpt-4-turbo-2024-04-09 $10.00 / 1M tokens $30.00 / 1M tokens gpt-3.5-turbo-0125 $0.50 / 1M tokens $1.50 / 1M tokens gpt-3.5-turbo-1106 $1.00 / 1M tokens $2.00 / 1M tokens gpt-4-1106-preview $10.00 / 1M tokens $30.00 / 1M tokens gpt-4 $30.00 / 1M tokens $60.00 / 1M tokens text-embedding-3-small $0.02 / 1M tokens text-embedding-3-large $0.13 / 1M tokens text-embedding-ada-0002 $0.10 / 1M tokens
Source code in mirascope/openai/utils.py
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 |
|
retry(fn)
¶
Decorator for retrying a function.