by Hunyuan
Hunyuan Hy3 preview is designed for agent workloads, adopting a MoE architecture with 295B capacity and 21B activated parameters. It provides three modes within the same model—no_think (ultra-fast response), think_low (fast thinking), and think_high (deep reasoning)—to accommodate different latency and depth requirements from high-frequency interactions to complex engineering tasks. On code benchmarks such as SWE-bench Verified it approaches the current state of the art, and its 256K context supports cross-file code refactoring and long-document analysis. It is suitable for developers who require reliable task completion while being sensitive to inference costs.
Hunyuan Hy3 preview is designed for agent workloads, adopting a MoE architecture with 295B capacity and 21B activated parameters. It provides three modes within the same model—no_think (ultra-fast response), think_low (fast thinking), and think_high (deep reasoning)—to accommodate different latency and depth requirements from high-frequency interactions to complex engineering tasks. On code benchmarks such as SWE-bench Verified it approaches the current state of the art, and its 256K context supports cross-file code refactoring and long-document analysis. It is suitable for developers who require reliable task completion while being sensitive to inference costs.
hy3-preview has a 256,000 token context window.
On AIHubMix, hy3-preview costs $0.17 per million input tokens and $0.57 per million output tokens. Cached input reads are billed at $0.05 per million tokens.
hy3-preview accepts text input.
hy3-preview supports tool calling, function calling, structured outputs, web, long context and thinking. Per-protocol parameter support is listed in the capability table on this page.
hy3-preview is available through the AIHubMix unified API. The API is OpenAI-compatible: point your OpenAI SDK at https://aihubmix.com/v1, use your AIHubMix API key, and set the model name to hy3-preview — no other code changes needed.
hy3-preview is developed by Hunyuan. AIHubMix aggregates it alongside models from other providers behind one API and one bill.
The Hy3 official version is honed for real-world business scenarios, using a…
Using the Hunyuan Sheng 3D 3.1 model, it can generate higher-precision and higher-quality…
Hunyuan-A13B-Instruct has 8 billion parameters and can match larger models by activating…
Hunyuan-MT-7B is a lightweight translation model with 7 billion parameters, designed to…
Use hy3-preview via the AIHubMix unified API — one interface for every major LLM.