Hy3 preview
Tencent· released 22 Apr 2026
tencent/hy3-preview- Input / 1M tokens
- $0.18
- Output / 1M tokens
- $0.6
Longest answer: 236K tokens
About
Hy3 preview is Tencent’s text model in the tencent family, released on 2026-04-22. It is an open-weights Mixture-of-Experts model designed for agentic workflows and production use. Reasoning can be configured as disabled, low, or high, and the model supports tools. It accepts text and returns text. Its context window is 262,144 tokens, with a maximum output of 235,929 tokens. For teams comparing per-token costs, input is $0.18 USD per 1M tokens and output is $0.60 USD per 1M tokens. The listed model ID is tencent/hy3-preview, and its Hugging Face ID is tencent/Hy3-preview. The official listing is on OpenRouter. These details make Hy3 preview relevant to builders considering an open-weights model with configurable reasoning and tool support for text-based agent workflows.
Who it is for
It suits teams building text-based agent workflows that want open weights, tool support, and configurable reasoning. It may also fit production-oriented deployments with large context needs.
What is good
- Open weights are available.
- Reasoning has disabled, low, and high modes.
- Tool support is listed.
- Context window is 262,144 tokens.
- Input is $0.18 USD per 1M tokens.
What to know first
- Text is the only listed input and output modality.
- Output costs $0.60 USD per 1M tokens.
Verdict
Hy3 preview pairs open weights and tool support with configurable reasoning and a 262,144-token context window. Teams should weigh its text-only modality and listed input and output rates against their workload.
Details
- Lab
- Tencentopenrouter.ai · 3 Oct 2026
- Context
- 262Kopenrouter.ai · 3 Oct 2026
- Input price
- $0.18 / 1Mopenrouter.ai · 3 Oct 2026
- Output price
- $0.6 / 1Mopenrouter.ai · 3 Oct 2026
- Max output
- 235,929 tokensopenrouter.ai · 3 Oct 2026
- Inputs
- textopenrouter.ai · 3 Oct 2026
- Open weights
- Yesopenrouter.ai · 3 Oct 2026
More newest models
See the list
Listed on Inferse
- Newest Models in 2026466 listed
- Longest-Context Models in 2026466 listed
- Cheapest Language Models in 2026437 listed
- Models with Tool Calling in 2026395 listed
- Reasoning Models in 2026332 listed
- Open-Weight Models in 2026178 listed
- Tencent Models: Pricing per 1M Tokens and Context (2026)7 listed
Sources
- openrouter.ai/tencent/hy3-preview· checked 3 Oct 2026