Built for agents
Tencent designed HY4 Preview for multi-step agent workflows, tool calling and long chains of execution.
Tencent
Operational· 100% uptime, 7 daysHY4 Preview API
The HY4 Preview API: Tencent's agent and coding model, 52% below Tencent's price
Price per 1M tokens52%off
- Input
- $0.40
$0.834 - Output
- $1.20
$2.501 - Cached input
- $0.02
Model ID
hy4-previewchat.py
from openai import OpenAIclient = OpenAI( base_url="https://api.zurelay.com/v1", api_key="YOUR_ZURELAY_KEY",)stream = client.chat.completions.create( model="hy4-preview", messages=[{"role": "user", "content": "Explain quicksort in two sentences."}], stream=True,)for chunk in stream: if chunk.choices: print(chunk.choices[0].delta.content or "", end="")Base URL https://api.zurelay.com/v1. Works with any OpenAI SDK.
- Context window
- 1M tokens
- Max output
- 64K tokens
- Input
- Text
- Output
- Text
- Released
- Aug 28, 2026
- Uptime, 7 days
- 100%
Pricing
HY4 Preview API pricing. A fraction of the list price.
Pay as you go from prepaid credit, with no subscription and no minimum. Requests that fail are never billed.
| Rate | zurelay | Tencent list | You save |
|---|---|---|---|
Input per 1M tokens | $0.40 | 52%off | |
Output per 1M tokens | $1.20 | 52%off | |
Cached input per 1M tokens | $0.02 | — | — |
One rate at every prompt length, up to the full 1M context window. Streaming, tool calls and structured outputs cost nothing extra.
Savings calculator
Move the sliders to your monthly usage.
Input tokens per month50M
Output tokens per month10M
zurelay
$32.00
Tencent list price
$66.71
You save every month$34.71
$416.52 a year
Overview
What is HY4 Preview?
HY4 Preview is Tencent's new HY model (formerly Hunyuan), with 770B total parameters and 49B active, built for agents, coding and productivity work. The zurelay HY4 Preview API serves it through one OpenAI-compatible endpoint with a 1M-token context window, at 52% below Tencent's official price.
Tencent released HY4 Preview on August 28, 2026, as an open-source model. HY is the new name for Tencent's Hunyuan models. HY4 Preview has 770B total parameters, of which 49B are active per token, and Tencent built it for agents, coding and productivity: multi-step agent workflows, tool calling, code generation and long chains of execution.
Tencent reports gains in long-context development, where it says the model is better at understanding, planning, debugging and validating code, and in office work such as financial analysis, data analysis and work across several documents. In an internal blind evaluation with 163 experts and 203 engineering tasks, Tencent says HY4 Preview averaged 2.99 out of 4, slightly ahead of GLM-5.3 at 2.92 and Kimi K3 at 2.94.
It is a reasoning model: it thinks before it answers, and that thinking counts toward the output budget. The context window is 1,048,576 tokens and a response can run to 64K output tokens, so give it a generous max_tokens on hard prompts, or the answer may be cut short after the reasoning.
On zurelay the model ID is hy4-preview, and tencent/hy4-preview and hy4 work as aliases. You pay $0.40 per 1M input tokens and $1.20 per 1M output tokens, with cache hits at $0.02, against Tencent's official price of $0.834 and $2.501.
Strengths
Where HY4 Preview shines. And what teams build with it.
Long-context coding
Tencent reports stronger understanding, planning, debugging and validation on long-context development tasks.
Office and analysis work
Tencent highlights complex working environments, financial and data analysis, and work that spans several documents.
1M tokens of context
1,048,576 tokens hold a large repository, a long agent trace or a stack of documents in one request.
Use cases
Coding agents
Plan, write, debug and validate code across a large repository, with tool calls in the loop.
Agent workflows
Run long chains of tool calls, such as research, data pulls and ticket handling, where the model has to keep track of many steps.
Financial and data analysis
Read reports and exported data, compare figures across several documents and draft the analysis.
Game prototypes
Turn a short game idea into a playable prototype. Tencent shows the model doing this from a single natural-language request.
Get started
Call HY4 Preview in three steps.
No waitlist and no new SDK. If your code already talks to OpenAI, it already talks to zurelay.
- 1
Create a key
Sign up, add credit and create an API key. Set a monthly budget or a rate limit per key if you like.
- 2
Point your SDK at zurelay
Change the base URL. Everything else in your code stays the same.
https://api.zurelay.com/v1 - 3
Use hy4-preview
Send chat completions as usual. Streaming, tool calls and usage reporting work as you expect.
hy4-preview
Works with the tools you already use
Compare
Similar models. All on the same key.
58%off
Kimi K3
$1.27 / $6.37 per 1M tokens
Pick Kimi K3 for long-horizon coding with image input; Tencent's own blind evaluation scored the two close together.
Kimi K3 API73%off
GLM-5.3
$0.38 / $1.21 per 1M tokens
Pick GLM-5.3 to compare another coding and agent model that Tencent's blind evaluation placed just behind HY4 Preview.
GLM-5.3 API40%off
Qwen 3.8 Max
$1.20 / $3.60 per 1M tokens
Pick Qwen 3.8 Max for Alibaba's larger flagship, with image and video input and a 1M-token context.
Qwen 3.8 Max APIFAQ
HY4 Preview API questions.
Anything else, write to sales@zurelay.com and an engineer will answer.
How much does the HY4 Preview API cost on zurelay?
On zurelay, hy4-preview costs $0.40 per 1M input tokens and $1.20 per 1M output tokens, with cache hits at $0.02. Tencent's official price is $0.834 input and $2.501 output, so you save 52%. Reasoning tokens bill as output.
Is there a cheaper HY4 Preview API than Tencent's?
Yes. zurelay serves hy4-preview at 52% below Tencent's official price. It is the same model, not a smaller substitute. You top up prepaid credit and pay only for the tokens you use, with no subscription.
How do I call the HY4 Preview API with the OpenAI SDK?
Set the OpenAI SDK's base URL to https://api.zurelay.com/v1, use your zurelay API key and pass model "hy4-preview". zurelay also accepts tencent/hy4-preview and hy4. Set max_tokens high enough for both the reasoning and the answer, since the model thinks before it replies.
What is the HY4 Preview context window?
HY4 Preview holds 1,048,576 tokens of context and returns up to 64K output tokens per response. Reasoning and the answer share that output budget, so leave headroom on hard prompts.
Is HY4 Preview a reasoning model?
Yes. It thinks before it answers, and those reasoning tokens count toward max_tokens and bill as output. Give it a generous max_tokens so the reasoning doesn't use up the budget before the answer starts.
What is HY4 Preview best at?
Tencent built it for agents, coding and productivity: multi-step agent workflows, tool calling, code generation and long chains of execution. Tencent also reports gains in office work, such as financial and data analysis, and on complex research problems.
Is Tencent HY the same as Hunyuan?
Yes. HY is the new name for Tencent's Hunyuan models, and HY4 Preview belongs to the HY4 series. Tencent released it as an open-source model on August 28, 2026.
Does the HY4 Preview API support streaming and per-key rate limits?
Yes. Set stream: true to receive tokens as they are generated, and add stream_options.include_usage to get token counts in the final chunk. Each zurelay API key can have its own monthly budget and requests-per-minute cap. Failed attempts are retried automatically and never billed.
Is HY4 Preview on zurelay the same model Tencent serves?
Yes. Requests go to hy4-preview itself, never a smaller or substitute model. Output is sampled, so wording varies from run to run, as it does on Tencent's own API. zurelay is an independent service and is not affiliated with or endorsed by Tencent.
More models on zurelay
All models- Claude Opus 5.5 API$1.40 / $7.00 per 1M tokens−65%
- GPT-6 Astra API$2.50 / $12.50 per 1M tokens−75%
- Nano Banana Pro APIfrom $0.035 per image−74%
- Nano Banana 2 APIfrom $0.025 per image−63%
- Claude Sonnet 5.5 API$0.80 / $4.00 per 1M tokens−60%
- GPT-6.1 Sol API$0.50 / $2.50 per 1M tokens−75%
- GPT-6 Sol API$0.50 / $2.50 per 1M tokens−75%
- GPT-6 Luna API$0.025 / $0.125 per 1M tokens−75%
- Grok 4.7 API$0.70 / $2.10 per 1M tokens−65%
- MiMo V2.6 Pro API$0.18 / $0.36 per 1M tokens−59%
- MiMo V2.6 Flash API$0.06 / $0.12 per 1M tokens−57%
- DeepSeek V4.1 Flash API$0.037 / $0.15 per 1M tokens−88%
- DeepSeek V4 Flash API$0.045 / $0.18 per 1M tokens−85%
- DeepSeek V4 Flash 0731 API$0.022 / $0.06 per 1M tokens−61%
- DeepSeek V4 Pro 0813 API$0.23 / $0.62 per 1M tokens−84%
- Claude Fable 5.1 API$3.50 / $17.50 per 1M tokens−65%
- Claude Fable 5 API$3.50 / $17.50 per 1M tokens−65%
- Claude Opus 5 API$1.75 / $8.75 per 1M tokens−65%
- Claude Opus 4.8 API$1.75 / $8.75 per 1M tokens−65%
- Claude Opus 4.7 API$1.78 / $8.93 per 1M tokens−64%
- Claude Opus 4.6 API$1.75 / $8.75 per 1M tokens−65%
- Claude Haiku 4.5 API$0.38 / $1.90 per 1M tokens−62%
- GPT-5.6 Sol API$1.14 / $5.70 per 1M tokens−72%
- GPT-5.6 Terra API$0.54 / $3.26 per 1M tokens−73%
- GPT-5.6 Luna API$0.08 / $0.49 per 1M tokens−59%
- Gemini 3.7 Flash API$0.32 / $1.59 per 1M tokens−58%
- Gemini 3.5 Flash-Lite API$0.12 / $1.04 per 1M tokens−58%
- Gemini 3.1 Flash-Lite API$0.10 / $0.59 per 1M tokens−61%
- GLM-5.3 API$0.38 / $1.21 per 1M tokens−73%
- GLM-5.2 API$0.38 / $1.21 per 1M tokens−73%
- Kimi K3 API$1.27 / $6.37 per 1M tokens−58%
- Qwen 3.8 Max API$1.20 / $3.60 per 1M tokens−40%
- Qwen 3.8 Flash API$0.07 / $0.22 per 1M tokens−53%
- Qwen 3.8 Omni Flash API$0.075 / $0.235 per 1M tokens−50%
- MiniMax M2.7 API$0.13 / $0.51 per 1M tokens−57%
- GPT Image 2.5 Flare API$0.02 per image−62%
- GPT Image 2.5 Sunburst API$0.02 per image−62%
- GPT Image 2 API$0.02 per image−62%
- Seedance 2.5 APIfrom $0.14 per secondVideo
- Seedance 2.0 APIfrom $0.097 per secondVideo
- Seedance 2.0 Fast APIfrom $0.074 per secondVideo
- Seedance 2.0 Mini APIfrom $0.047 per secondVideo
Stop paying list price.
Start saving today.
Create a key in seconds, point the SDK you already use at zurelay, and every request costs up to 85% less from the first token.