Skip to content
NEWNano Banana 2.1 is live — Google's latest image model, from $0.0336 per image, native 4K, about half the price of the previous generationView model
HopBase
← Back to blog

Hunyuan Hy4 API Pricing and Quickstart (hy4-preview) | HopBase

HopBase Nov 5, 2026 9 min read

The short answer: Tencent Hunyuan Hy4 Preview (model ID hy4-preview) is live on HopBase. It is the preview release of Tencent's fourth-generation Hunyuan model: a mixture-of-experts model with about 770B total and 49B active parameters, a 1M-token (1,048,576) context window and up to 64K output tokens. That makes it a good fit for long-document Q&A, repository-scale code analysis and agent workflows that call tools. Call it with POST /v1/chat/completions, or use /v1/responses or the Anthropic-format /v1/messages. It is billed at the official list price (table below); as of November 5, 2026 the Tencent Hunyuan Hy4 group is billed at 67% of list price, all day — the group description in the console is authoritative.

How to call it

  • Base URL: https://api.hop-base.com/v1 for OpenAI-compatible SDKs; only the base URL and API key change.
  • Use the full model ID: hy4-preview, as returned by GET /v1/models.
  • Endpoints: POST /v1/chat/completions (recommended), POST /v1/responses, and POST /v1/messages for Anthropic-format tools such as Claude Code.
  • Groups: any key whose groups include Tencent Hunyuan Hy4 can call it; otherwise add that group to the key in the console.

Pricing: official list price

ItemTencent Hunyuan Hy4 Preview
Input$0.8824 / 1M tokens
Cached input$0.0441 / 1M tokens
Output$2.65 / 1M tokens

This is Tencent Hunyuan's official list price — ¥6 input, ¥18 output and ¥0.3 cached input per 1M tokens — converted at 6.8 CNY per USD. You are billed list price × actual usage × your group's rate; at the current 67% group rate that is roughly $0.59 input and $1.78 output per 1M tokens. Repeated long prefixes (system prompts, long documents) hit the cache and are billed at the cached-input price. For volume pricing, contact sales from the pricing page.

Specs and limits

  • Context window: 1,048,576 tokens
  • Max output: 64K tokens
  • Tool calling: supported; streaming tool calls verified on Chat Completions, Responses and Messages
  • Input: text only — image input is not supported
  • Status: vendor preview; the name or specs may change, so check GET /v1/models

Three-minute quickstart

curl https://api.hop-base.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hy4-preview",
    "messages": [{"role": "user", "content": "Introduce yourself in three sentences"}]
  }'
from openai import OpenAI

client = OpenAI(base_url="https://api.hop-base.com/v1", api_key="sk-your-key")
resp = client.chat.completions.create(
    model="hy4-preview",
    messages=[{"role": "user", "content": "Introduce yourself in three sentences"}],
)
print(resp.choices[0].message.content)

Full parameters and examples: Tencent Hunyuan Hy4 Preview model docs. New to HopBase? Create an account. If you already have a key, call GET /v1/models to confirm hy4-preview is listed.

FAQ

How do I call Tencent Hunyuan Hy4 Preview?

Use POST /v1/chat/completions with model ID hy4-preview and base URL https://api.hop-base.com/v1. With the OpenAI SDK you only change the base URL and API key; Anthropic-format tools such as Claude Code use /v1/messages.

How much does Tencent Hunyuan Hy4 Preview cost?

The official list price is $0.8824 input, $2.65 output and $0.0441 cached input per 1M tokens (¥6 / ¥18 / ¥0.3). As of November 5, 2026 the group is billed at 67% of list all day; your effective rate is shown on your group in the console.

How long is the context, and can it read images?

The context window is 1M tokens with up to 64K output tokens. Input is text only; images are not supported.

Product