Hy3 preview
Tencent's Hy3 preview is a Mixture-of-Experts model engineered for agentic workflows and production environments. It features configurable reasoning levels and supports a large context window for processing extensive text data.
Hy3 preview specs and pricing
- Provider: Tencent
- Released: April 22, 2026
- Context window: 262K tokens
- Max output: 236K tokens
- Input: text
- Output: text
- Price: $0.18 input / $0.60 output per 1M tokens
- Reasoning: yes
- Tool calling: yes
What Hy3 preview is good at
- Complex agentic workflows requiring multi-step reasoning
- Processing large-scale text documents
- Automated tasks that require tool calling capabilities
To-dos to try with Hy3 preview
- Analyze a 200,000-token technical manual to identify specific system requirements.
- Execute a multi-step research task using the model's tool-calling functionality.
- Draft a comprehensive project report based on a large dataset using the high reasoning mode.
FAQ
- What is the primary design focus of the Hy3 preview model? The model is specifically designed for agentic workflows and production use cases.
- Does the model support different levels of reasoning? Yes, it offers configurable reasoning levels including disabled, low, and high modes.
- How much does Hy3 preview cost? Hy3 preview costs $0.18 per million input tokens and $0.60 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 0.59 credits ($1 = 100 credits).
- What is the context window of Hy3 preview? Hy3 preview can read up to 262K tokens at once and write up to 236K tokens in one reply.
- Can I use Hy3 preview without coding? Yes. In todo.is you pick Hy3 preview as your agent's model and write to-dos in plain words, on the web, the Android app, WhatsApp or Telegram. No API key needed: credits are included, or you can bring your own key.
JavaScript is required to use the todo.is app.