Fireworks AI
The Fireworks AI provider uses provider key fireworks and the HPD-Agent.Providers.Fireworks package. ModelName is a Fireworks chat model id.
The provider package is currently net10.0 only and uses HPD Agent's shared OpenAI-compatible chat-completions client.
Set an API key:
export FIREWORKS_API_KEY="..."The default endpoint is https://api.fireworks.ai/inference/v1/. Pass endpoint to WithFireworks(...), set Endpoint in ChatClientConfig, or use FIREWORKS_ENDPOINT / FIREWORKS_BASE_URL when you need a compatible proxy or custom Fireworks base URL.
Use fluent chat setup first:
using HPD.Agent;
using HPD.Agent.Providers.Fireworks;
var agent = await new AgentBuilder()
.WithFireworks(model: "accounts/fireworks/models/llama-v3p1-8b-instruct")
.BuildAsync();
var result = await agent.RunAsync("Write one sentence about Fireworks AI setup.");
Console.WriteLine(result.Text);The equivalent Clients.Chat shape is:
{
"Clients": {
"Chat": {
"ProviderKey": "fireworks",
"ModelName": "accounts/fireworks/models/llama-v3p1-8b-instruct"
}
}
}Runtime Chat Options
Use ChatClientConfig for shared model-call behavior such as temperature, top-p, max output tokens, stop sequences, response format, tools, tool mode, seed, and generic reasoning options. Put agent-level defaults in AgentConfig.Clients.Chat; use AgentRunConfig.Clients.Chat for complete per-run replacements.
Supported ResponseFormat values are text and json_object. Supported ToolChoice values are auto, none, and required.
Caveats
This package registers chat only. Fireworks audio/STT, embeddings, image generation, and hosted-file families are not registered by this provider.
Runtime validation checks model and endpoint configuration. Runtime chat options are applied through Clients.Chat and AgentRunConfig.Clients.Chat. Credentials are resolved lazily when a run first acquires this provider client; a missing credential fails that run, not BuildAsync().