Models / Tencent
Hy4-preview
by Tencent
Used for frontier, coding and reasoning.
Best forCodingReasoningFrontier
Hy4-preview is made by Tencent. It takes text in and answers in text, with a context window of 1M tokens.
Price per 1M tokens
The live price, read from billing as this page loads. You pay only for the tokens you use, from prepaid credit.
Tīrtha Bench review
What people use it for, in its maker's words:
Try it
Model id catalog/hy4-preview · base URL https://api.tirtha.ai/v1. Any OpenAI-compatible client works: change the base URL and keep your code.
Using a coding tool? Set up OpenCode, Cline, Cursor and more. PowerShell and the full reference: Quickstart.
Limits and support
- Text inYes
- ReasoningYesThinks before it answers
- StreamingYes
- Sampling settingsYou can set themThe maker suggests temperature 0.9, top p 1, top k -1.
What it is for
- Coding
Software engineering: Better at understanding, planning, debugging, and verifying long-horizon development tasks
From the maker's card - Reasoning
Stronger understanding, reasoning, and problem-solving on hard research questions
From the maker's card - Frontier
Hy4 preview is a new-generation Mixture-of-Experts (MoE) flagship model
From the maker's card
In the maker's own words, from its published model card. Claims and benchmarks on that card are the maker's own, not measured by Tīrtha.
What a job costs
1,000 calls of 2,000 tokens in and 500 out cost about -
Worked out from the live price above, without cached input. Your own calls will differ in size.
Similar models on Tīrtha
Licence
Weights: Apache-2.0. Read the licence.
From the maker: Model card: tencent/Hy4-preview on Hugging Face · Last updated by the maker: Aug 28, 2026Facts from the maker's card are the maker's own, not measured by Tīrtha.