11-19 更新 实时
首页/ 科技公司/ 正文

腾讯发布 Hy4 flash,内部测试任务成本降 95%

科技公司 来源:TechNode 2026-11-16 14:54 6 阅读 机器翻译
腾讯发布 Hy4 flash,内部测试任务成本降 95%

腾讯周一发布了 Hy4 flash,这是一款成本更低的 AI 模型,面向研究、制作幻灯片和编辑电子表格等日常工作。在腾讯内部的 WorkIndex 基准测试中,Hy4 flash 每项任务的成本是完整版 Hy4 模型的二十分之一。API 标价为每百万输入 Token 0.07 美元、每百万输出 Token 0.281 美元、每百万缓存输入 Token 0.003 美元。

Hy4 flash 支持多模态输入,上下文窗口最高可达 100 万 Token。它有 2240 亿参数,每次前向传播激活 120 亿参数。在另一项盲测中,85 名 WorkBuddy 用户对 259 项任务的输出进行了评分,腾讯称该模型在用户满意度上接近 Claude Haiku 5.5,并且在受测模型中成本最低、完成时间最短。

该模型已在 WorkBuddy 和元宝中上线,也可通过腾讯云的 TokenHub 使用。WorkBuddy 用户可在 11 月 30 日前免费使用,TokenHub 则以标价的一半提供 API 访问。腾讯计划发布该模型的权重,并使其在 OpenRouter 上可用。[TechNode 报道]

译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。

查看英文原文(Tencent launches Hy4 flash, cutting AI task costs by 95% in internal tests)

Tencent on Monday launched Hy4 flash, a lower-cost AI model for everyday work such as research, slide creation and spreadsheet editing. On Tencent’s internal WorkIndex benchmark, Hy4 flash cost one-twentieth as much per task as the full Hy4 model. API list prices are $0.07 per million input tokens, $0.281 per million output tokens and $0.003 per million cached input tokens.

Hy4 flash supports multimodal inputs and a context window of up to one million tokens. It has 224 billion parameters and activates 12 billion per forward pass. In a separate blind test in which 85 WorkBuddy users rated outputs across 259 tasks, Tencent said the model scored close to Claude Haiku 5.5 in user satisfaction and had the lowest cost and shortest completion time among the models tested.

The model is available in WorkBuddy and Yuanbao and through Tencent Cloud’s TokenHub. WorkBuddy users can access it for free through November 30, and TokenHub is offering API access at half the list price. Tencent plans to release the model’s weights and make it available on OpenRouter. [TechNode reporting]

正文保留英文原文,标题与摘要由机器翻译。

本文由采集器自动同步

原文发布于 TechNode(Tencent launches Hy4 flash, cutting AI task costs by 95% in internal tests),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。

阅读原文 ↗