# Hy4 Flash now available on AI Gateway for 50% off

Source: [Vercel](https://vercel.com/changelog/hy4-flash-now-available-on-ai-gateway-for-50-off)  
Feed7 permalink: https://feed7.dev/p/hy4-flash-now-available-on-ai-gateway-for-50-off-01mclxb  
Published: 2026-11-16T17:00:00.000Z  
Trust: Official Source (official_source)

## Why Included

Hy4 Flash gives coding agents a lower-cost option with tool calling, prompt caching, image input, and a 1M-token context window; AI Gateway’s 50% discount ends November 29, 2026.

## Source Summary

Tencent’s **Hy4 Flash** is available through AI Gateway at about one tenth of the Hy4 family’s per-token price. It supports tool calling, structured outputs, prompt caching, image input, and a **1M-token context window**; reasoning can be disabled. A **50% introductory discount** runs through November 29, 2026.
For high-volume coding or document tasks, try it on workloads where cost matters and compare results with your current model. The model ID is `tencent/hy4-flash`; Vercel says it can be configured for Claude Code, Codex, and Cursor through its CLI.
The discount is temporary, and the material gives no benchmark results. Measure quality and cost on your own agent tasks before switching.

## Practical Implication

No separate implication recorded.

## Agent-Ready Context

Tencent’s **Hy4 Flash** is available through AI Gateway at about one tenth of the Hy4 family’s per-token price. It supports tool calling, structured outputs, prompt caching, image input, and a **1M-token context window**; reasoning can be disabled. A **50% introductory discount** runs through November 29, 2026.
For high-volume coding or document tasks, try it on workloads where cost matters and compare results with your current model. The model ID is `tencent/hy4-flash`; Vercel says it can be configured for Claude Code, Codex, and Cursor through its CLI.
The discount is temporary, and the material gives no benchmark results. Measure quality and cost on your own agent tasks before switching.

## Connected Context

Feed7 judgment across 1116 accumulated Signals:

Hy4 Flash adds a lower-cost Hy4 route for long-context coding and document work, with tool calling, structured outputs, caching, and image input. The temporary discount and absence of benchmark results make this a workload-trial candidate, not evidence to change a default. Its listed capabilities do not include the computer-use support described for Hy4.

- [Tencent Hy4 now available on AI Gateway](https://feed7.dev/p/tencent-hy4-now-available-on-ai-gateway-1encpce) — Hy4 Flash is positioned as the cheaper Hy4 variant; the Hy4 candidate adds vision and computer use, making task modality a concrete distinction to test alongside price.
- [Gemini 3.8 Flash now available on AI Gateway](https://feed7.dev/p/gemini-3-8-flash-now-available-on-ai-gateway-0cun77q) — Both are discounted, million-token coding-agent options without comparative results, reinforcing the need to compare them on matched workloads and separate promotional cost from lasting fit.

## Context Map

- Layer: model
- Domains: coding
- Topics: model-selection, coding-agents, context-caching

## Uncertainty

- None recorded.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.
