OpenAI (Independent Publisher) (Preview)
Connect to the OpenAI API and use the Power of GPT3, API key must be entered as "Bearer YOUR_API_KEY"
This connector is available in the following products and regions:
| Service | Class | Regions |
|---|---|---|
| Copilot Studio | Premium | All Power Automate regions except the following: - US Government (GCC) - US Government (GCC High) - China Cloud operated by 21Vianet - US Department of Defense (DoD) |
| Logic Apps | Standard | All Logic Apps regions except the following: - Azure Government regions - Azure China regions - US Department of Defense (DoD) |
| Power Apps | Premium | All Power Apps regions except the following: - US Government (GCC) - US Government (GCC High) - China Cloud operated by 21Vianet - US Department of Defense (DoD) |
| Power Automate | Premium | All Power Automate regions except the following: - US Government (GCC) - US Government (GCC High) - China Cloud operated by 21Vianet - US Department of Defense (DoD) |
| Contact | |
|---|---|
| Name | Robin Rosengrün |
| URL | https://linktr.ee/r2power |
| robin@r2power.de |
| Connector Metadata | |
|---|---|
| Publisher | Robin Rosengrün |
| Website | https://openai.com/ |
| Privacy policy | https://openai.com/api/policies/terms/ |
| Categories | AI |
Creating a connection
The connector supports the following authentication types:
| Default | Parameters for creating connection. | All regions | Not shareable |
Default
Applicable: All regions
Parameters for creating connection.
This is not shareable connection. If the power app is shared with another user, another user will be prompted to create new connection explicitly.
| Name | Type | Description | Required |
|---|---|---|---|
| API Key | securestring | Enter API Key as as "Bearer YOUR_API_KEY" | True |
Throttling Limits
| Name | Calls | Renewal Period |
|---|---|---|
| API calls per connection | 100 | 60 seconds |
Actions
| Chat Completion |
Use models like ChatGPT and GPT4 to hold a conversation |
| Create an Image |
DallE2 creates an image from your prompt |
| Embeddings |
Get a vector representation of a given input |
| GPT3 Completes your prompt |
GPT3 Completes your prompt |
| GPT3 Completes your prompt [DEPRECATED] |
GPT3 Completes your prompt (deprecated by OpenAI - use Completion_New) |
Chat Completion
Use models like ChatGPT and GPT4 to hold a conversation
Parameters
| Name | Key | Required | Type | Description |
|---|---|---|---|---|
|
model
|
model | True | string |
The used model, choose between gpt-3.5-turbo, gpt-4 and others |
|
role
|
role | True | string |
The role of the author of this message. One of system, user, or assistant. |
|
content
|
content | True | string |
The contents of the message. |
|
n
|
n | integer |
How many completions to generate for each prompt |
|
|
temperature
|
temperature | float |
Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer. Use this OR top p |
|
|
max tokens
|
max_tokens | integer |
One token equals roughly 4 characters of text (up to 4000 or more tokens between prompt and completion, depending on model) |
|
|
top p
|
top_p | float |
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. |
|
|
frequency penalty
|
frequency_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the models likelihood to repeat the same line verbatim. |
|
|
presence penalty
|
presence_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the models likelihood to talk about new topics. |
|
|
stop
|
stop | array of string |
Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence |
Returns
| Name | Path | Type | Description |
|---|---|---|---|
|
id
|
id | string |
id |
|
object
|
object | string |
object |
|
created
|
created | integer |
created |
|
choices
|
choices | array of object |
choices |
|
index
|
choices.index | integer |
index |
|
role
|
choices.message.role | string |
role |
|
content
|
choices.message.content | string |
content |
|
finish_reason
|
choices.finish_reason | string |
finish_reason |
|
prompt_tokens
|
usage.prompt_tokens | integer |
prompt_tokens |
|
completion_tokens
|
usage.completion_tokens | integer |
completion_tokens |
|
total_tokens
|
usage.total_tokens | integer |
total_tokens |
Create an Image
DallE2 creates an image from your prompt
Parameters
| Name | Key | Required | Type | Description |
|---|---|---|---|---|
|
prompt
|
prompt | True | string |
The prompt that describes the image |
|
Number of images
|
n | integer |
Number of images from 1 to 10 |
|
|
size
|
size | string |
The size of the generated images. 256x256, 512x512 or 1024x1024 (default: 1024x1024) |
|
|
format
|
response_format | string |
Get url to picture or receive it in base64 format (default: url) |
Returns
| Name | Path | Type | Description |
|---|---|---|---|
|
data
|
data | array of object |
data |
|
url
|
data.url | string |
URL to created Image |
|
b64image
|
data.b64_json | byte |
Image in base64 format |
Embeddings
Get a vector representation of a given input
Parameters
| Name | Key | Required | Type | Description |
|---|---|---|---|---|
|
model
|
model | True | string |
model |
|
input
|
input | True | string |
input |
Returns
| Name | Path | Type | Description |
|---|---|---|---|
|
object
|
object | string |
object |
|
data
|
data | array of object |
data |
|
object
|
data.object | string |
object |
|
embedding
|
data.embedding | array of float |
embedding |
|
index
|
data.index | integer |
index |
|
model
|
model | string |
model |
|
prompt_tokens
|
usage.prompt_tokens | integer |
prompt_tokens |
|
total_tokens
|
usage.total_tokens | integer |
total_tokens |
GPT3 Completes your prompt
GPT3 Completes your prompt
Parameters
| Name | Key | Required | Type | Description |
|---|---|---|---|---|
|
Engine
|
model | True | string |
The used model, choose between text-davinci-002, text-curie-001, text-babbage-001, text-ada-001 |
|
prompt
|
prompt | True | string |
Text that will be completed by GPT3 |
|
n
|
n | integer |
How many completions to generate for each prompt |
|
|
best_of
|
best_of | integer |
If set to more than 1, generates multiple completions server-side and returns the "best". Must be greater than "n". Use with caution, can consume a lot of tokens. |
|
|
temperature
|
temperature | float |
Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer. Use this OR top p |
|
|
max tokens
|
max_tokens | integer |
One token equals roughly 4 characters of text (up to 4000 tokens between prompt and completion) |
|
|
top p
|
top_p | float |
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. |
|
|
frequency penalty
|
frequency_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the models likelihood to repeat the same line verbatim. |
|
|
presence penalty
|
presence_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the models likelihood to talk about new topics. |
|
|
stop
|
stop | array of string |
Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence |
Returns
| Name | Path | Type | Description |
|---|---|---|---|
|
id
|
id | string |
id |
|
object
|
object | string |
object |
|
created
|
created | integer |
created |
|
choices
|
choices | array of object |
Returned Completion(s) |
|
Text
|
choices.text | string |
Completion text |
|
Index
|
choices.index | integer |
Number of completion |
|
Finish reason
|
choices.finish_reason | string |
Reason why the text finished (stop condition / natural end / length) |
|
Prompt Tokens
|
choices.usage.prompt_tokens | integer |
Number of tokens in the prompt |
|
Completion Tokens
|
choices.usage.completion_tokens | integer |
Number of tokens in the completion |
|
Total Tokens
|
choices.usage.total_tokens | integer |
Total number of tokens in prompt and completion |
GPT3 Completes your prompt [DEPRECATED]
GPT3 Completes your prompt (deprecated by OpenAI - use Completion_New)
Parameters
| Name | Key | Required | Type | Description |
|---|---|---|---|---|
|
Engine
|
engine | True | string |
The used engine, choose between text-davinci-002/003, text-curie-001, text-babbage-001, text-ada-001 |
|
prompt
|
prompt | True | string |
Text that will be completed by GPT3 |
|
n
|
n | integer |
How many completions to generate for each prompt |
|
|
best_of
|
best_of | integer |
If set to more than 1, generates multiple completions server-side and returns the "best". Must be greater than "n". Use with caution, can consume a lot of tokens. |
|
|
temperature
|
temperature | float |
Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer. Use this OR top p |
|
|
max tokens
|
max_tokens | integer |
One token equals roughly 4 characters of text (up to 4000 tokens between prompt and completion) |
|
|
top p
|
top_p | float |
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. |
|
|
frequency penalty
|
frequency_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the models likelihood to repeat the same line verbatim. |
|
|
presence penalty
|
presence_penalty | float |
Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the models likelihood to talk about new topics. |
|
|
user
|
user | string |
A unique identifier representing your end-user, which will help OpenAI to monitor and detect abuse |
|
|
stop
|
stop | array of string |
Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence |
Returns
| Name | Path | Type | Description |
|---|---|---|---|
|
id
|
id | string |
id |
|
object
|
object | string |
object |
|
created
|
created | integer |
created |
|
choices
|
choices | array of object |
Returned Completion(s) |
|
Text
|
choices.text | string |
Completion text |
|
Index
|
choices.index | integer |
Number of completion |
|
Logprobs
|
choices.logprobs | string |
Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens. For example, if logprobs is 3, the API will return a list of the 3 most likely tokens. |
|
Finish reason
|
choices.finish_reason | string |
Reason why the text finished (stop condition / natural end / length) |