LLM Reference
Tencent Cloud TI Platform

Hunyuan Hy3 Preview on Tencent Cloud TI Platform

Hunyuan · Tencent AI Lab

ServerlessOpen Weights

Last refreshed 2026-06-07. Next refresh: weekly.

Why use Hunyuan Hy3 Preview on Tencent Cloud TI Platform?

Tencent Cloud TI Platform offers Hunyuan Hy3 Preview with pay-as-you-go pricing at $0.07/1M input tokens. Tencent Cloud's AI Platform offers a comprehensive suite of innovative solutions leveraging artificial intelligence to drive business growth and shape a more intelligent future.

Compare Hunyuan Hy3 Preview across 2 providers to find the best fit for your use case
Input / 1M
$0.066
Output / 1M
$0.26
Cache
Not sourced
Batch
Not sourced

Setup recipe

Docs fallback
Install
Use the provider REST API or SDK
Auth
Create a provider API key
Call
model: tencent/hy3-preview
Model ID
tencent/hy3-preview

Request example

Curated snippets for this provider are not sourced yet. Use Tencent Cloud TI Platform documentation with model ID tencent/hy3-preview.

Gotchas

  • Use provider model ID "tencent/hy3-preview", not the LLMReference slug "hy3-preview".

Compare Hunyuan Hy3 Preview Across Providers

ProviderInput (per 1M)Output (per 1M)
OpenRouter$0.07$0.26
Tencent Cloud TI Platform$0.07$0.26

Pricing

TypePrice (per 1M)
Input tokens$0.07
Output tokens$0.26

Capabilities

ReasoningJSON / Tool use

About Hunyuan Hy3 Preview

Tencent HunYuan's Hy3 Preview is a high-efficiency Mixture-of-Experts language model for agentic and production workflows. OpenRouter lists tencent/hy3-preview as released Apr 22, 2026 with 262,144 context, preview pricing, reasoning controls, and tool-use support. Hugging Face Transformers documents Hy3-preview as a Tencent HunYuan MoE model with a dense-MoE hybrid architecture, 192 routed experts, and one always-active shared expert per MoE layer. SCMP's Apr 23 coverage reports Tencent described HY3-Preview as a new flagship model developed by the HunYuan and Yuanbao teams with 295B parameters.

Get Started

Model Specs

Released2026-04-22
Parameters295B
Context262k
ArchitectureDense + MoE Hybrid