# 腾讯发布开源MoE模型Hy3，性能声称匹敌5倍规模

- 来源：The Decoder：AI News（RSS）
- 作者：Matthias Bastian
- 发布时间：2026-07-07 02:03
- AIHOT 分数：70
- AIHOT 标记：同事件
- AIHOT 链接：https://aihot.virxact.com/items/cmr9k8mdn00hlihe8imji93dl
- 原文链接：https://the-decoder.com/tencent-releases-hy3-open-source-model-that-allegedly-matches-models-up-to-five-times-its-active-size

## AI 摘要

腾讯正式发布开源MoE模型Hy3，总参数295B，活跃参数21B，另有3.8B MTP层，上下文窗口256k token。官方称其性能可比肩两到五倍规模模型。270位专家盲评中Hy3得分2.67/4，超越GLM-5.1的2.51；内部测试幻觉率从12.5%降至5.4%。Hy3以Apache 2.0许可在Hugging Face、ModelScope和GitHub开源，提供FP8量化版，计划支持OpenRouter和Cline。腾讯已将其集成至WorkBuddy、元宝、微信及游戏《Path of Exile: Advent》等产品。

## 正文

Tencent has officially released its AI model Hy3. The model uses a Mixture-of-Experts (MoE) architecture with 295 billion total parameters. Of those, 21 billion are active at any given time, plus 3.8 billion parameters for an added MTP layer. It handles context lengths up to 256,000 tokens. Tencent says Hy3 matches the performance of models two to five times its size. In a blind evaluation by 270 experts, Hy3 scored 2.67 out of 4, beating GLM-5.1 at 2.51. Internal testing showed the hallucination rate dropped from 12.5 percent to 5.4 percent.

Hy3 benchmark results compared to other models. | Image: Tencent

Hy3 is available under an Apache 2.0 license on Hugging Face, ModelScope, and GitHub. An FP8-quantized version is also available. Support for platforms like OpenRouter and Cline is planned. Tencent has already built the model into its own products, including WorkBuddy, Yuanbao, WeChat, and the game assistant for "Path of Exile: Advent."