Darkspyre.Inference.Connect.OpenAiApi 11.19.3-rc.2

Darkspyre.Inference.Connect.OpenAiApi

Conversations, agent sessions and text embeddings through the OpenAI Responses API, at OpenAI or at another service that serves the same API at its own endpoint, such as a local model server. Its provider appears as "OpenAI API Key".

Install

dotnet add package Darkspyre.Inference.Connect.OpenAiApi

Register

services.AddLlmConnect();

AddLlmConnect() from Darkspyre.Inference.Connect.Common discovers this package and registers its conversation client, its agent sessions and its embedding client.

Example

var connection = new LlmConnection
{
    Version = version,
    Transport = LlmTransport.OpenAiResponses,
    Endpoint = "https://api.openai.com/v1",
    Credential = protectedApiKeyCredentialJson
};

LlmEmbeddingProviderService embeddings = serviceProvider.GetRequiredService<LlmEmbeddingProviderService>();
LlmEmbeddingResult result = await embeddings.EmbedAsync(connectionId, new LlmEmbeddingRequest
{
    ModelId = "text-embedding-3-small",
    Inputs = [new TextEmbeddingInput("The meeting moved to Thursday.")]
}, cancellationToken);

The application returns the connection from its ILlmConnectionSource. Conversations open through LlmAgentProviderService.StartSessionAsync.

Settings

• Endpoint is the API's base address. It defaults to https://api.openai.com/v1.
• API key is optional. Leave it blank for a service that takes no key.
• Service tiers shows the service-tier choice in the chat panel and sends the chosen tier. It starts checked on api.openai.com and unchecked on any other endpoint.
• Context lengths rows each set one model's context length.
• Check authentication and model access loads the model list from the endpoint.

Limits

• An OpenAI API key comes from an OpenAI platform account and is billed separately from a ChatGPT subscription. A ChatGPT subscription alone is not an API key; use Darkspyre.Inference.Connect.Codex for one.
• When the endpoint refuses temperature or reasoning settings for a model, the failure names the refused setting, so the application can send the request again without it.

The Darkspyre.Inference packages

This package is one library of the Darkspyre.Inference family. The family runs AI inference from .NET applications through typed, provider-neutral contracts.

• Darkspyre.Inference.Common holds the contracts every other package shares: requests, results, capabilities, failures and the provider setup model.
• Darkspyre.Inference.Connect.Common holds the conversation and embedding services. Each Connect provider package adds one provider, and AddLlmConnect() discovers every provider package the application references.
• The Tensor packages load and run models inside the application's own process.
• The UI.Settings packages provide the setup editors a host places in its own window or page.

Every package of the family carries the same version. Reference all of them at the same version: each package depends on the exact version of the family packages it uses.

The packages are licensed under the MIT license. LICENSE.md in each package holds the license text.