Streaming Chat Inference

What problem does it solve?

This skill orchestrates verified 0G Compute chat providers to deliver streamed or standard conversational responses while ensuring provider acknowledgement, usage tracking, and fee settlement.

Core Features & Use Cases

• Streaming and non-streaming modes: Works with DeepSeek, Qwen, Gemma, or GPT-OSS models to stream tokens or return completed responses.
• Provider lifecycle management: Handles broker initialization, metadata retrieval, auth headers, and usage reporting that prevents settlement failures.
• Use Case: Deploy a customer-facing chatbot that streams replies, captures chat IDs from headers, and calls processResponse after each request so provider fees settle automatically.

Quick Start

Use streaming chat inference to send a user prompt through the broker, collect the chat id from headers, and immediately invoke processResponse with the reported usage.