Updated

Hunyuan API

Tencent's Hunyuan (HY) models through Tencent Cloud's TokenHub, with OpenAI- and Anthropic-compatible endpoints in Singapore, the US and mainland China.

Paid from Pay per token

Profile checked

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗
Recent changesFollow Hunyuan API (RSS)
  • Oct 2026 Now covered: Hunyuan API, for llm api.

How makers use Hunyuan API

What Hunyuan API is used for

LLM API · Model provider →Tencent's HY models plus DeepSeek, GLM, Kimi and MiniMax on one TokenHub key, in OpenAI and Anthropic formats.Trade-off: HY vision and translation models are only on the mainland site, and processing can leave your chosen region for operational reasons.

Hunyuan API vs the alternatives makers chose

OptionMakers' products2026 launchesFree forAt 10,000 usersOpen source
OpenAI API736 80%77%~1,000 users$10/mo · GPT-6 LunaNo
Claude252 27%21%~1,000 users$10/mo · Claude Haiku 5.5No
Gemini API99 11%12%~360 users$28/mo · Gemini 3.1 Flash-LiteNo
Groq39 4%3%~1,500 users$6.75/mo · GPT OSS 20BNo
Azure OpenAI22 2%0%——No
Mistral AI15 2%1%~1,700 users$6.00/mo · Ministral 3 (3B)No
Hunyuan API0 0%0%~840 users—No

For LLM API: how many of the 923 makers' products we know the LLM API of chose each option, and the share among the 256 launched in 2026, since a tool that came first has products from every year, and what each costs at 10,000 users.

Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.

When to use Hunyuan API, and when not

Something else fits better if

  • You want the strongest general models and the widest ecosystem of examples and integrations: OpenAI API or Claude or Gemini API.
  • Token cost dominates your budget, for example high-volume batch or agent workloads: DeepSeek or Groq or Mistral AI.
  • You want to try or mix many models with one key and one bill, and fall back when a provider is down: OpenRouter.
  • You want that routing and spend control but self-hosted, with keys and logs on your servers: LiteLLM.
From each decision's guide, written from what makers' products chose and why.

The Hunyuan API is how you call Tencent's Hunyuan models, now branded HY: the Hy3 language model and a Hy4 preview, plus vision and translation models on the mainland site. Internationally they're sold through TokenHub, Tencent Cloud's model service, which also serves DeepSeek, GLM, Kimi and MiniMax models under the same account. Tencent's older standalone Hunyuan console on its mainland China cloud no longer takes new purchases.

More about Hunyuan API

Every TokenHub model accepts both the OpenAI Chat Completions format and the Anthropic Messages format, so you use the official OpenAI or Anthropic SDKs by swapping the base URL and key. Each region has its own endpoint: https://tokenhub-intl.tencentcloudmaas.com/v1 for Singapore, https://tokenhub-us.tencentcloudmaas.tech/v1 for Silicon Valley, and https://tokenhub.tencentcloudmaas.com/v1 for Guangzhou in mainland China. Hy3 is called as hy3.

Useful for a small team: one key covers Tencent's models and several other Chinese model families; Hy3 is published as open weights on Hugging Face under Apache 2.0, so you can serve it yourself; and Token Plan subscriptions are sold alongside pay-per-token billing.

Limits: Tencent, based in Shenzhen, processes data in the service region you select, but its data processing agreement lets it process data outside that region for operational, regulatory or other reasons. Prices also differ by region, so choose one before you build.

Who uses it

No maker's live product on record yet.

We haven't found a product that uses Hunyuan API yet.

Alternatives to Hunyuan API

All alternatives by situation →

Further reading

Questions makers ask about Hunyuan API

What are the alternatives to Hunyuan API?

For LLM API, the other options makers' products we track chose most are OpenAI API (736), Claude (252), Gemini API (99), Groq (39), Azure OpenAI (22), Mistral AI (15). Each alternative, with the situation it fits, is on the Hunyuan API alternatives page.

Which endpoints and SDKs can I use?

TokenHub accepts the OpenAI Chat Completions and Anthropic Messages formats for all models. OpenAI-style calls send Authorization Bearer with your key, Anthropic-style calls send x-api-key, against the Singapore, Silicon Valley or Guangzhou base URL. source ↗

Where is my data processed?

Under Tencent Cloud's data processing agreement, if you select a service region Tencent processes personal data in that region, though the agreement allows processing outside it for operational, regulatory or other reasons. Prices also differ by region, so choose one before you build. source ↗

What happens when I hit a rate limit?

TokenHub returns a 429-series error code, sometimes with a Retry-After header, and Tencent recommends retrying with exponential backoff. source ↗

Which other models are on TokenHub?

Besides Tencent's own HY models, TokenHub serves models from DeepSeek, MiniMax, Kimi and Zhipu GLM through the same API. source ↗

Can I run Hunyuan myself?

Hy3 is published by Tencent on Hugging Face under the Apache 2.0 license, so you can serve it on your own hardware. source ↗

How much does Hunyuan API cost?

Paid use starts at Pay per token. source ↗

Is Hunyuan API open source or self-hostable?

Not open source, and hosted only.

Can AI coding agents work with Hunyuan API?

No llms.txt, official MCP server or CLI found yet.

Who uses Hunyuan API?

No tracked product yet. source ↗