Intel / article

Tencent Hy4 now available on AI Gateway

Source: vercel.com
Author: Jerilyn Zheng
Date: Published Nov 3, 2026

Context

Tencent's Hy4 model can now be called through Vercel's AI Gateway, a service that gives developers one API for many models and handles usage and cost tracking, retries and failover. Vercel's changelog lists the model at 50% off through December 31, 2026, and says the gateway passes through provider pricing with no markup and no platform fee, including when customers bring their own provider keys. Developers can also pick it inside coding agents such as Claude Code, Codex and Cursor after running the Vercel CLI's setup command.

The clearest change from Hy4 Preview, which Vercel added to the gateway earlier, is vision. Vercel describes Hy4 as the first Hy model that can see images, able to pick out fine detail and read complex charts, and says it supports computer use, meaning it can operate both command-line and graphical interfaces. It also lists reasoning, , structured outputs and a 1M- , and says Hy4 follows instructions more closely than the preview and does better with long context, long-running tasks and deep research. The changelog gives no results and no standard price, so those comparisons are Vercel's description, not measured gains.

Key takeaways · AI-distilled

Hy4 is Tencent's first Hy model with vision: it accepts image input, reads complex charts, and supports computer use in both command-line and graphical interfaces.
Hy4 supports reasoning, tool calling and structured outputs with a 1M-token context window; Vercel says it follows instructions more closely than Hy4 Preview and does better on long-context, long-running and deep research tasks.
On Vercel AI Gateway, Hy4 is 50% off through December 31, 2026, and the gateway passes through provider pricing with no markup, including on Bring Your Own Key requests.

Terms in this piece · Glossary

tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.

Read the source vercel.com