From 6d649f1bb1dfda53483b217ef7de77a9fcc5890f Mon Sep 17 00:00:00 2001 From: sr-remsha <46220620+sr-remsha@users.noreply.github.com> Date: Thu, 29 Aug 2024 16:55:41 +0200 Subject: [PATCH] chore: Update README.md (#462) --- README.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 86378a76f..1938b226e 100644 --- a/README.md +++ b/README.md @@ -211,19 +211,19 @@ Dynamic settings can include the following parameters: | assistant | A list of deployed AI DIAL Assistants and their parameters:
``: Unique assistan name. | | assistant.endpoint | Assistant main endpoint | | assistant.assistants. | `iconUrl`: Icon path for the AI DIAL assistant on UI.
`description`: Brief AI DIAL assistant description.
`displayName`: AI DIAL assistant name on UI.
`inputAttachmentTypes`: A list of allowed MIME types for the input attachments.
`maxInputAttachments`: Maximum number of input attachments (default is zero when `inputAttachmentTypes` is unset, otherwise, infinity)
`forwardAuthToken`: If flag is set to `true` forward Http header with authorization token to chat completion endpoint of the assistant.
`userRoles`: a specific claim value provided by a specific IDP. Refer to [IDP Configuration](https://github.com/epam/ai-dial/blob/main/docs/Auth/2.%20Web/1.overview.md) to view examples. | -| assistant.assistants..defaults | default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | +| assistant.assistants..defaults | Default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | | applications | A list of deployed AI DIAL Applications and their parameters:
``: Unique application name. | | applications. | `endpoint`: AI DIAL Application API for chat completions.
`iconUrl`: Icon path for the AI DIAL Application on UI.
`description`: Brief AI DIAL Application description.
`displayName`: AI DIAL Application name on UI.
`inputAttachmentTypes`: A list of allowed MIME types for the input attachments.
`maxInputAttachments`: Maximum number of input attachments (default is zero when `inputAttachmentTypes` is unset, otherwise, infinity)
`forwardAuthToken`: If flag is set to `true` forward Http header with authorization token to chat completion endpoint of the application.
`userRoles`: a specific claim value provided by a specific IDP. Refer to [IDP Configuration](https://github.com/epam/ai-dial/blob/main/docs/Auth/2.%20Web/1.overview.md) to view examples. | -| applications..defaults | default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | -| models..interceptors | list of interceptors to be triggered for the given application | +| applications..defaults | Default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | +| applications..interceptors | A list of interceptors to be triggered for the given application | | models | A list of deployed models and their parameters:
``: Unique model name. | -| models. | `type`: Model type—`chat` or `embedding`.
`iconUrl`: Icon path for the model on UI.
`description`: Brief model description.
`displayName`: Model name on UI.
`displayVersion`: Model version on UI.
`endpoint`: Model API for chat completions or embeddings.
`tokenizerModel`: Identifies the specific model whose tokenization algorithm exactly matches that of the referenced model. This is typically the name of the earliest-released model in a series of models sharing an identical tokenization algorithm (e.g. `gpt-3.5-turbo-0301`, `gpt-4-0314`, or `gpt-4-1106-vision-preview`). This parameter is essential for DIAL clients that reimplement tokenization algorithms on their side, instead of utilizing the `tokenizeEndpoint` provided by the model.
`features`: Model features.
`limits`: Model token limits.
`pricing`: Model pricing.
`upstreams`: Used for load-balancing—request is sent to model endpoint containing X-UPSTREAM-ENDPOINT and X-UPSTREAM-KEY headers.
`userRoles`: a specific claim value provided by a specific IDP. Refer to [IDP Configuration](https://github.com/epam/ai-dial/blob/main/docs/Auth/2.%20Web/1.overview.md) to view examples. | +| models. | `type`: Model type—`chat` or `embedding`.
`iconUrl`: Icon path for the model on UI.
`description`: Brief model description.
`displayName`: Model name on UI.
`displayVersion`: Model version on UI.
`endpoint`: Model API for chat completions or embeddings.
`tokenizerModel`: Identifies the specific model whose tokenization algorithm exactly matches that of the referenced model. This is typically the name of the earliest-released model in a series of models sharing an identical tokenization algorithm (e.g. `gpt-3.5-turbo-0301`, `gpt-4-0314`, or `gpt-4-1106-vision-preview`). This parameter is essential for DIAL clients that reimplement tokenization algorithms on their side, instead of utilizing the `tokenizeEndpoint` provided by the model.
`features`: Model features.
`limits`: Model token limits.
`pricing`: Model pricing.
`upstreams`: Used for [load-balancing—request](https://docs.epam-rail.com/tutorials/load-balancer) is sent to model endpoint containing X-UPSTREAM-ENDPOINT and X-UPSTREAM-KEY headers.
`userRoles`: a specific claim value provided by a specific IDP. Refer to [IDP Configuration](https://github.com/epam/ai-dial/blob/main/docs/Auth/2.%20Web/1.overview.md) to view examples. | | models..limits | `maxPromptTokens`: maximum number of tokens in a completion request.
`maxCompletionTokens`: maximum number of tokens in a completion response.
`maxTotalTokens`: maximum number of tokens in completion request and response combined.
Typically either `maxTotalTokens` is specified or `maxPromptTokens` and `maxCompletionTokens`. | | models..pricing | `unit`: the pricing units (currently `token` and `char_without_whitespace` are supported).
`prompt`: per-unit price for the completion request in USD.
`completion`: per-unit price for the completion response in USD. | | models..features | `rateEndpoint`: endpoint for rate requests *(exposed by core as `/rate`)*.
`tokenizeEndpoint`: endpoint for requests to the model tokenizer *(exposed by core as `/tokenize`)*.
`truncatePromptEndpoint`: endpoint for truncating prompt requests *(exposed by core as `/truncate_prompt`)*.
`systemPromptSupported`: does the model support system prompt (default is `true`).
`toolsSupported`: does the model support tools (default is `false`).
`seedSupported`: does the model support `seed` request parameter (default is `false`).
`urlAttachmentsSupported`: does the model/application support attachments with URLs (default is `false`).
`folderAttachmentsSupported`: does the model/application support folder attachments (default is `false`) | -| models..upstreams | `endpoint`: Model endpoint.
`key`: Your API key.
`weight`: Weight for upstream endpoint; positive number represents an endpoint capacity, zero or negative disables this enpoint from routing. Default value: 1.
`tier`: Specifies tier group for the endpoint. Only positive numbers allowed. All requests will be routed to the endpoints with highest tier (lowest tier value), other endpoints (with lower tier/higher tier value) may be used if and only if highest tier endpoints are unavailable. Default value: 0 - highest tier | -| models..defaults | default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | -| models..interceptors | list of interceptors to be triggered for the given model | +| models..upstreams | `endpoint`: Model endpoint.
`key`: Your API key.
`weight`: Weight for upstream endpoint; positive number represents an endpoint capacity, zero or negative disables this enpoint from routing. Default value: 1.
`tier`: Specifies tier group for the endpoint. Only positive numbers allowed. All requests will be routed to the endpoints with the highest tier (the lowest tier value), other endpoints (with lower tier/higher tier value) may be used only if the highest tier endpoints are unavailable. Default value: 0 - highest tier. Refer to [Load Balancer](https://docs.epam-rail.com/tutorials/load-balancer) to learn more. | +| models..defaults | Default parameters are applied if a request doesn't contain them in OpenAI `chat/completions` API call | +| models..interceptors | A list of interceptors to be triggered for the given model | | keys | API Keys parameters:
``: Your API key. Refer to [API Keys](https://github.com/epam/ai-dial/blob/main/docs/Roles%20and%20Access%20Control/3.API%20Keys.md) to learn more. | | keys. | `project`: Project name is assigned to this key.
`role`: A configured role name to be defined in the section `roles`. **Note**: both attributes are required.
`secured`: the flag indicates if the key is secured. If it's set to `true` user request and deployment response won't be saved to the prompt log storage. | | roles | API key or user roles. Each role may have limits to be associated with applications, models, assistants or addons. Refer to [API Keys](https://github.com/epam/ai-dial/blob/main/docs/Roles%20and%20Access%20Control/3.API%20Keys.md) to learn more. |