Hy MT2 30B A3B
Neural Network
Hy MT2 30B A3B is Tencent's flagship translation model from the Hy-MT2 family for glossary-based and context-aware translation.
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Hy MT2 30B A3B?
Hy MT2 30B A3B is Tencent's flagship translation model in the Hy-MT2 family. It covers 33 language pairs plus five more with Chinese dialects and minority languages, and supports structured translation, delimiter handling, context awareness, and glossary usage. It is designed for tasks where terminology consistency across the entire text is more important than individual phrases. In BotHub, the model is accessible without a VPN or foreign card; payment is made with Russian cards on a pay-as-you-go basis, with no subscription required. Access over 250 models in one window with a unified OpenAI-compatible API, so you won't need to rewrite integrations when switching. For companies, we offer contracts, invoicing, electronic document management, and an admin panel with limits. It is useful for localizing interfaces and documentation, translating client correspondence and support, preparing materials for multiple markets simultaneously, maintaining a project terminology database, or embedding translation into your own product via API.Frequently asked questions about Hy MT2 30B A3B
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
hy-mt2-30b-a3b is a Tencent text model: it works with text and accepts documents as input. It is suitable for analyzing and summarizing files, editing, drafting, and answering questions based on uploaded material. The context window is 8192 tokens, so it is best suited for short to medium-length documents.
The model generates up to 4096 tokens per response, which is several pages of text. If you need extensive material, break the task into parts: ask for an outline first, then the sections one by one. Keep in mind that the total dialogue context is limited to 8192 tokens.
Our data does not explicitly list a reasoning mode for hy-mt2-30b-a3b, but that does not mean it lacks one. You can get a step-by-step breakdown using a standard prompt: ask it to break down the task into steps and explain the logic before the final conclusion. The easiest way to verify this is to test it yourself.
Support for function calling and strict JSON output is not explicitly listed in our data—we won't deny it, but we cannot promise it either. If a schema is mandatory, describe the required format directly in the prompt or switch to a model with confirmed function calling: you won't need to rewrite the unified BotHub API.
In addition to text, the model accepts documents—this is confirmed by our data. Support for images, audio, and video is not listed, nor is generation of anything other than text. For images, audio, or video, switch to a specialized model in the same BotHub window—no separate subscription is required.
We do not have data on the quality of performance for specific languages, so we cannot provide an assessment—it is easier to check in practice. Send a couple of your typical tasks and compare the response with other models: switching in BotHub takes seconds, and you only pay for the tokens used.