Side-by-side comparison of Hy3-preview (Tencent) and Kimi K2 Instruct (Moonshot AI · China) for self-hosted deployment of the open-weight model. Hy3-preview is rated blocked; Kimi K2 Instruct is conditional. They part ways on licence: Hy3-preview is "Tencent Hy Community", Kimi K2 Instruct is "Modified MIT".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | Blocked Per the Tencent Hy Community License Agreement (Sections 1(l) and 5(c)), the licence's defined Territory excludes the European Union, the United Kingdom and South Korea, and licensees are expressly prohibited from using, distributing or displaying the model or its outputs outside that Territory. Under those terms the weights are not deployable for EU users or workloads, regardless of architecture quality. | Conditional 1T-parameter MoE (32B active) tuned for agentic and tool-use workflows. Modified MIT permits commercial use. Same Chinese-origin alignment and supply-chain considerations as DeepSeek and Qwen. |
| Last reviewed | 2026-04-28 | 2026-04-16 |
| Open-weight | ||
| Licence | Tencent Hy Community | Modified MIT |
| Commercial use | EU territory excluded | Yes |
| Training data | Undisclosed | Undisclosed |
| Origin | China | China |
| Performance & pricing? | ||
| Quality index | 42/100 | 26/100 |
| Speed | 85 tok/s | 34 tok/s |
| Blended price | — | $1.04/M |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.