What is Hy3 API?
Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It offers configurable reasoning effort, reliable tool-calling across agent frameworks, and strong performance on coding, document analysis, finance, and frontend development.
Technical Specifications
- Supported inputs: Text prompt (chat messages)
- Output: Text, with configurable reasoning effort
- Context window: 262,144 tokens / Max output: 131,072 tokens
- Endpoint: /v1/chat/completions
Performance Benchmarks
[embed]
Hy3 API Pricing
- Input: $0.182 / 1M tokens
- Cached input: $0.0455 / 1M tokens
- Output: $0.754 / 1M tokens
Code Sample
[embed]
Comparison with Other Models
- vs other reasoning models: Hy3 combines configurable reasoning effort with a large 295B MoE architecture at a competitive price point.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
Hy3 is documented as suited to software development, where it shows progress in code generation and agent capabilities and is used in products like CodeBuddy and WorkBuddy for development workflows. It is also suited to office productivity, including document generation and workflow tasks such as creating PowerPoint, Word, Excel, PDF, and HTML files.
Hy3 is documented for financial modeling, where Tencent says it performs well in productivity scenarios alongside coding and office work. It is also suited to front-end design and game production, which are specifically named as productivity scenarios where the model has made notable progress.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "tencent/hy3", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "tencent/hy3", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"tencent/hy3","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 25.8 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Coding | 58.8 | Composite score across standardised coding evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Hy3 This page | Reasoning + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
Hy3 has a 262,144 tokens context window and can return up to 131,072 tokens.
Hy3 takes image, text as input and returns text.
Use tencent/hy3 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
Hy3 became available on July 6, 2026.
Hy3 is priced at input $0.27508 / 1M tokens, output $1.10032 / 1M tokens, cached input $0.06877 / 1M tokens.
Yes, Hy3 can stream responses as they are generated.
Yes, Hy3 accepts image input alongside text.
Hy3 was built by Tencent.
