View Source OpenAI (openai.ex v0.5.2)
Provides API wrappers for OpenAI API See https://beta.openai.com/docs/api-reference/introduction for further info on REST endpoints
Link to this section Summary
Functions
Transcribes audio into the input language.
Translates audio into into English.
Creates a completion for the chat message
It returns one or more predicted completions given a prompt. The function accepts as arguments the "engine_id" and the set of parameters used by the Completions OpenAI api
It returns one or more predicted completions given a prompt. The function accepts as arguments the "engine_id" and the set of parameters used by the Completions OpenAI api
Creates a new edit for the provided input, instruction, and parameters
Creates an embedding vector representing the input text.
@deprecated: "use models instead" Get the list of available engines
Returns a list of files that belong to the user's organization.
delete a file
Upload a file that contains document(s) to be used across various endpoints/features. Currently, the size of all the files uploaded by one organization can be up to 1 GB. Please contact OpenAI if you need to increase the storage limit.
List your organization's fine-tuning jobs
Immediately cancel a fine-tune job.
Creates a job that fine-tunes a specified model from a given dataset.
Delete a fine-tuned model. You must have the Owner role in your organization.
Get fine-grained status updates for a fine-tune job.
alias of images_edits(file_path, params, request_options) - will be deprecated in future releases
alias of images_generations(params, request_options) - will be deprecated in future releases
alias of images_variations(file_path, params, request_options) - will be deprecated in future releases
This edits an image based on the given prompt. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
This generates an image based on the given prompt. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
Creates a variation of a given image. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
Retrieve the list of available models
Classifies if text violates OpenAI's Content Policy
Callback implementation for Application.start/2
.
Link to this section Functions
Transcribes audio into the input language.
example-request
Example request
OpenAI.audio_transcription( "./path_to_file/blade_runner.mp3", # file path model: "whisper-1" )
example-response
Example response
{:ok, %{ text: "I've seen things you people wouldn't believe.." }}
See: https://platform.openai.com/docs/api-reference/audio/create
Translates audio into into English.
example-request
Example request
OpenAI.audio_translation( "./path_to_file/werner_herzog_interview.mp3", # file path model: "whisper-1" )
example-response
Example response
{:ok, %{ text: "I thought if I walked, I would be saved. It was almost like a pilgrimage. I will definitely continue to walk long distances. It is a very unique form of life and existence that we have lost almost entirely from our normal life." }}
See: https://platform.openai.com/docs/api-reference/audio/create
Creates a completion for the chat message
example-request
Example request
OpenAI.chat_completion(
model: "gpt-3.5-turbo",
messages: [
%{role: "system", content: "You are a helpful assistant."},
%{role: "user", content: "Who won the world series in 2020?"},
%{role: "assistant", content: "The Los Angeles Dodgers won the World Series in 2020."},
%{role: "user", content: "Where was it played?"}
]
)
example-response
Example response
{:ok,
%{
choices: [
%{
"finish_reason" => "stop",
"index" => 0,
"message" => %{
"content" => "The 2020 World Series was played at Globe Life Field in Arlington, Texas due to the COVID-19 pandemic.",
"role" => "assistant"
}
}
],
created: 1677773799,
id: "chatcmpl-6pftfA4NO9pOQIdxao6Z4McDlx90l",
model: "gpt-3.5-turbo-0301",
object: "chat.completion",
usage: %{
"completion_tokens" => 26,
"prompt_tokens" => 56,
"total_tokens" => 82
}
}
}
N.B. to use "stream" mode you must be set http_options as below when you want to treat the chat completion as a stream:
config :openai, api_key: "your-api-key", http_options: [recv_timeout: :infinity, stream_to: self(), async: :once]
example-request-stream
Example request (stream)
OpenAI.chat_completion([
model: "gpt-3.5-turbo",
messages: [
%{role: "system", content: "You are a helpful assistant."},
%{role: "user", content: "Who won the world series in 2020?"},
%{role: "assistant", content: "The Los Angeles Dodgers won the World Series in 2020."},
%{role: "user", content: "Where was it played?"}
],
stream: true, # set this param to true
]
)
|> Stream.each(fn res ->
IO.inspect(res)
end)
|> Stream.run()
example-response-stream
Example response (stream)
%{
"choices" => [
%{"delta" => %{"role" => "assistant"}, "finish_reason" => nil, "index" => 0}
],
"created" => 1682700668,
"id" => "chatcmpl-7ALbIuLju70hXy3jPa3o5VVlrxR6a",
"model" => "gpt-3.5-turbo-0301",
"object" => "chat.completion.chunk"
}
%{
"choices" => [
%{"delta" => %{"content" => "The"}, "finish_reason" => nil, "index" => 0}
],
"created" => 1682700668,
"id" => "chatcmpl-7ALbIuLju70hXy3jPa3o5VVlrxR6a",
"model" => "gpt-3.5-turbo-0301",
"object" => "chat.completion.chunk"
}
%{
"choices" => [
%{"delta" => %{"content" => " World"}, "finish_reason" => nil, "index" => 0}
],
"created" => 1682700668,
"id" => "chatcmpl-7ALbIuLju70hXy3jPa3o5VVlrxR6a",
"model" => "gpt-3.5-turbo-0301",
"object" => "chat.completion.chunk"
}
%{
"choices" => [
%{
"delta" => %{"content" => " Series"},
"finish_reason" => nil,
"index" => 0
}
],
"created" => 1682700668,
"id" => "chatcmpl-7ALbIuLju70hXy3jPa3o5VVlrxR6a",
"model" => "gpt-3.5-turbo-0301",
"object" => "chat.completion.chunk"
}
%{
"choices" => [
%{"delta" => %{"content" => " in"}, "finish_reason" => nil, "index" => 0}
],
"created" => 1682700668,
"id" => "chatcmpl-7ALbIuLju70hXy3jPa3o5VVlrxR6a",
"model" => "gpt-3.5-turbo-0301",
"object" => "chat.completion.chunk"
}
See: https://platform.openai.com/docs/api-reference/chat/create for the complete list of parameters you can pass to the completions function
It returns one or more predicted completions given a prompt. The function accepts as arguments the "engine_id" and the set of parameters used by the Completions OpenAI api
example-request
Example request
OpenAI.completions(
model: "finetuned-model",
prompt: "once upon a time",
max_tokens: 5,
temperature: 1,
...
)
example-response
Example response
{:ok, %{
choices: [
%{
"finish_reason" => "length",
"index" => 0,
"logprobs" => nil,
"text" => "" thing we are given"
}
],
created: 1617147958,
id: "...",
model: "...",
object: "text_completion"
}
} See: https://platform.openai.com/docs/api-reference/completions/create
It returns one or more predicted completions given a prompt. The function accepts as arguments the "engine_id" and the set of parameters used by the Completions OpenAI api
example-request
Example request
OpenAI.completions(
"davinci", # engine_id
prompt: "once upon a time",
max_tokens: 5,
temperature: 1,
...
)
example-response
Example response
{:ok, %{
choices: [
%{
"finish_reason" => "length",
"index" => 0,
"logprobs" => nil,
"text" => "" thing we are given"
}
],
created: 1617147958,
id: "...",
model: "...",
object: "text_completion"
}
}
See: https://beta.openai.com/docs/api-reference/completions/create for the complete list of parameters you can pass to the completions function
Creates a new edit for the provided input, instruction, and parameters
example-request
Example request
OpenAI.edits( model: "text-davinci-edit-001", input: "What day of the wek is it?", instruction: "Fix the spelling mistakes" )
example-response
Example response
{:ok, %{ choices: [%{"index" => 0, "text" => "What day of the week is it? "}], created: 1675443483, object: "edit", usage: %{ "completion_tokens" => 28, "prompt_tokens" => 25, "total_tokens" => 53 } }}
See: https://platform.openai.com/docs/api-reference/edits/create
Creates an embedding vector representing the input text.
example-request
Example request
OpenAI.embeddings( model: "text-embedding-ada-002", input: "The food was delicious and the waiter..." )
example-response
Example response
{:ok, %{ data: [ %{
"embedding" => [0.0022523515000000003, -0.009276069000000001,
0.015758524000000003, -0.007790373999999999, -0.004714223999999999,
0.014806155000000001, -0.009803046499999999, -0.038323310000000006,
-0.006844355, -0.028672641, 0.025345700000000002, 0.018145794000000003,
-0.0035904291999999997, -0.025498080000000003, 5.142790000000001e-4,
-0.016317246, 0.028444072, 0.0053713582, 0.009631619999999999,
-0.016469626, -0.015390275, 0.004301531, 0.006984035499999999,
-0.007079272499999999, -0.003926933, 0.018602932000000003, 0.008666554,
-0.022717162999999995, 0.011460166999999997, 0.023860006,
0.015568050999999998, -0.003587254600000001, -0.034843990000000005,
-0.0041555012999999995, -0.026107594000000005, -0.02151083,
-0.0057618289999999996, 0.011714132499999998, 0.008355445999999999,
0.004098358999999999, 0.019199749999999998, -0.014336321, 0.008952264,
0.0063395994, -0.04576447999999999, ...],
"index" => 0,
"object" => "embedding"
} ], model: "text-embedding-ada-002-v2", object: "list", usage: %{"prompt_tokens" => 8, "total_tokens" => 8} }}
See: https://platform.openai.com/docs/api-reference/embeddings/create
@deprecated: "use models instead" Get the list of available engines
example-request
Example request
OpenAI.engines()
example-response
Example response
{:ok, %{
"data" => [
%{"id" => "davinci", "object" => "engine", "max_replicas": ...},
...,
...
]
}
See: https://beta.openai.com/docs/api-reference/engines/list
Retrieve specific engine info
example-request-1
Example request
OpenAI.engines("davinci")
example-response-1
Example response
{:ok, %{
"id" => "davinci",
"object" => "engine",
"max_replicas": ...
}
}
See: https://beta.openai.com/docs/api-reference/engines/retrieve
Returns a list of files that belong to the user's organization.
example-request
Example request
OpenAI.files()
example-response
Example response
{:ok,
%{
data: [
%{
"bytes" => 123,
"created_at" => 213,
"filename" => "file.jsonl",
"id" => "file-123321",
"object" => "file",
"purpose" => "fine-tune",
"status" => "processed",
"status_details" => nil
}
],
object: "list"
}
}
See: https://platform.openai.com/docs/api-reference/files
Returns a file that belong to the user's organization, given a file id
example-request-1
Example Request
OpenAI.files("file-123321")
example-response-1
Example Response
{:ok,
%{
bytes: 923,
created_at: 1675370979,
filename: "file.jsonl",
id: "file-123321",
object: "file",
purpose: "fine-tune",
status: "processed",
status_details: nil
}
}
See: https://platform.openai.com/docs/api-reference/files/retrieve
delete a file
example-request
Example Request
OpenAI.files_delete("file-123")
example-response
Example Response
{:ok, %{deleted: true, id: "file-123", object: "file"}}
See: https://platform.openai.com/docs/api-reference/files/delete
Upload a file that contains document(s) to be used across various endpoints/features. Currently, the size of all the files uploaded by one organization can be up to 1 GB. Please contact OpenAI if you need to increase the storage limit.
example-request
Example request
OpenAI.files_upload("./file.jsonl", purpose: "fine-tune")
example-response
Example response
{:ok,
%{
bytes: 923,
created_at: 1675373519,
filename: "file.jsonl",
id: "file-123",
object: "file",
purpose: "fine-tune",
status: "uploaded",
status_details: nil
}
}
See: https://platform.openai.com/docs/api-reference/files/upload
List your organization's fine-tuning jobs
example-request
Example request
OpenAI.finetunes()
example-response
Example response
{:ok, %{
"data" => [
%{"created_at" => 1614807352, "fine_tuned_model" => "curie:ft-acmeco-2021-03-03-21-44-20", "model": ...},
...,
...
]
}
See: https://beta.openai.com/docs/api-reference/fine-tunes/list
Gets info about the fine-tune job.
example-request-1
Example request
OpenAI.finetunes("ft-AF1WoRqd3aJAHsqc9NY7iL8F")
example-response-1
Example response
{:ok, %{
created_at: 1614807352,
events: [
%{
"created_at" => 1614807352,
"level" => "info",
"message" => "Created fine-tune: ft-AF1WoRqd3aJAHsqc9NY7iL8F",
"object" => "fine-tune-event"
},
%{
"created_at" => 1614807360,
"level" => "info",
"message" => "Fine-tune costs $0.02",
"object" => "fine-tune-event"
},
...,
...
}
See: https://beta.openai.com/docs/api-reference/fine-tunes/retrieve
Immediately cancel a fine-tune job.
example-request
Example request
OpenAI.finetunes_cancel("ft-AF1WoRqd3aJAHsqc9NY7iL8F")
example-response
Example response
{:ok, %{ created_at: 1675527767, events: [ ... %{
"created_at" => 1675528080,
"level" => "info",
"message" => "Fine-tune cancelled",
"object" => "fine-tune-event"
} ], fine_tuned_model: nil, hyperparams: %{ "batch_size" => 1, "learning_rate_multiplier" => 0.1, "n_epochs" => 4, "prompt_loss_weight" => 0.01 }, id: "ft-IaBYfSSAK47UUCbebY5tBIEj", model: "curie", object: "fine-tune", organization_id: "org-1iPTOIak4b5fpuIB697AYMmO", result_files: [], status: "cancelled", training_files: [ %{
"bytes" => 923,
"created_at" => 1675373519,
"filename" => "file123.jsonl",
"id" => "file-123",
"object" => "file",
"purpose" => "fine-tune",
"status" => "processed",
"status_details" => nil
} ], updated_at: 1675528080, validation_files: [] }}
See: https://platform.openai.com/docs/api-reference/fine-tunes/cancel
Creates a job that fine-tunes a specified model from a given dataset.
example-request
Example request
OpenAI.finetunes_create(training_file: "file-123", model: "curie", validation_file: "file-456")
example-response
Example response
See: https://platform.openai.com/docs/api-reference/fine-tunes/create
Delete a fine-tuned model. You must have the Owner role in your organization.
example-request
Example request
OpenAI.finetunes_delete_model("model-id")
example-response
Example response
{:ok, %{ id: "model-id", object: "model", deleted: true }}
See: https://platform.openai.com/docs/api-reference/fine-tunes/delete-model
Get fine-grained status updates for a fine-tune job.
example-request
Example request
OpenAI.finetunes_list_events("ft-AF1WoRqd3aJAHsqc9NY7iL8F")
example-response
Example response
{:ok, %{ data: [ %{
"created_at" => 1675376995,
"level" => "info",
"message" => "Created fine-tune: ft-123",
"object" => "fine-tune-event"
}, %{
"created_at" => 1675377104,
"level" => "info",
"message" => "Fine-tune costs $0.00",
"object" => "fine-tune-event"
}, %{
"created_at" => 1675377105,
"level" => "info",
"message" => "Fine-tune enqueued. Queue number: 18",
"object" => "fine-tune-event"
}, ..., ] } }
See: https://platform.openai.com/docs/api-reference/fine-tunes/events
alias of images_edits(file_path, params, request_options) - will be deprecated in future releases
alias of images_generations(params, request_options) - will be deprecated in future releases
alias of images_variations(file_path, params, request_options) - will be deprecated in future releases
This edits an image based on the given prompt. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
example-request
Example Request
OpenAI.images_edits(
"/home/developer/myImg.png",
[prompt: "A developer writing a test", "size": "256x256"],
%OpenAI.config{http_options: [recv_timeout: 10 * 60 * 1000]} # optional!
)
example-response
Example Response
{:ok,
%{
created: 1670341737,
data: [
%{
"url" => ...Returned url
}
]
}}
See: https://beta.openai.com/docs/api-reference/images/create-edit for the complete list of parameters you can pass to the image creation function note: the official way of passing http_options changed in v0.5.0 to be compliant with the conventions of other APIs, the alias OpenAI.images_edits(file_path, params, request_options), but is still available for retrocompatibility. If you are using it consider to switch to OpenAI.images_edits(file_path, params, config)
This generates an image based on the given prompt. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
example-request
Example Request
OpenAI.images_generations(
[prompt: "A developer writing a test", size: "256x256"],
%OpenAI.config{http_options: [recv_timeout: 10 * 60 * 1000]} # optional!
)
example-response
Example Response
{:ok,
%{
created: 1670341737,
data: [
%{
"url" => ...Returned url
}
]
}} See: https://beta.openai.com/docs/api-reference/images/create for the complete list of parameters you can pass to the image creation function
note: the official way of passing http_options changed in v0.5.0 to be compliant with the conventions of other APIs, the alias OpenAI.images_generations(file_path, params, request_options), but is still available for retrocompatibility. If you are using it consider to switch to OpenAI.images_variations(params, config)
Creates a variation of a given image. Image functions require some times to execute, and API may return a timeout error, if needed you can pass a configuration object with HTTPoison http_options as second argument of the function to increase the timeout.
example-request
Example Request
OpenAI.images_variations(
"/home/developer/myImg.png",
[n: "5"],
%OpenAI.config{http_options: [recv_timeout: 10 * 60 * 1000]} # optional!
)
example-response
Example Response
{:ok,
%{
created: 1670341737,
data: [
%{
"url" => ...Returned url
}
]
}}
See: https://beta.openai.com/docs/api-reference/images/create-variation for the complete list of parameters you can pass to the image creation function note: the official way of passing http_options changed in v0.5.0 to be compliant with the conventions of other APIs, the alias OpenAI.images_variations(file_path, params, request_options), but is still available for retrocompatibility. If you are using it consider to switch to OpenAI.images_edits(file_path, params, config)
Retrieve the list of available models
example-request
Example request
OpenAI.models()
example-response
Example response
%{
"created" => 1651172505,
"id" => "davinci-search-query",
"object" => "model",
"owned_by" => "openai-dev",
"parent" => nil,
"permission" => [
%{
"allow_create_engine" => false,
"allow_fine_tuning" => false,
"allow_logprobs" => true,
...
}
],
"root" => "davinci-search-query"
} See: https://platform.openai.com/docs/api-reference/models/retrieve
Retrieve specific model info
example-request-1
Example request
OpenAI.models("davinci-search-query")
example-response-1
Example response
{:ok, %{
data: [%{ "created" => 1651172505, "id" => "davinci-search-query", "object" => "model", "owned_by" => "openai-dev", "parent" => nil, "permission" => [
%{
"allow_create_engine" => false,
"allow_fine_tuning" => false,
"allow_logprobs" => true,
...
}
], "root" => "davinci-search-query" }, ....], object: "list" }} See: https://platform.openai.com/docs/api-reference/models/retrieve
Classifies if text violates OpenAI's Content Policy
example-request
Example request
OpenAI.moderations(input: "I want to kill everyone!")
example-response
Example response
{:ok, %{ id: "modr-6gEWXyuaU8dqiHpbAHIsdru0zuC88", model: "text-moderation-004", results: [ %{
"categories" => %{
"hate" => false,
"hate/threatening" => false,
"self-harm" => false,
"sexual" => false,
"sexual/minors" => false,
"violence" => true,
"violence/graphic" => false
},
"category_scores" => %{
"hate" => 0.05119025334715844,
"hate/threatening" => 0.00321022979915142,
"self-harm" => 7.337320857914165e-5,
"sexual" => 1.1111642379546538e-6,
"sexual/minors" => 3.588798147546868e-10,
"violence" => 0.9190407395362855,
"violence/graphic" => 1.2791929293598514e-7
},
"flagged" => true
} ] }}
See: https://platform.openai.com/docs/api-reference/moderations/create
Callback implementation for Application.start/2
.