腾讯混元新出的Hy3，295B参数的MoE模型，Agent任务比GLM-5.1强不少，还带256K上下文和FP8，开源随便玩。
腾讯混元在ModelScope发布Hy3模型，总参数量295B，21B激活的MoE架构，支持256K上下文和FP8量化版本。在人类评估中Hy3得分为2.67/4，超过GLM-5.1的2.51/4，前端开发、CI/CD和数据存储场景提升最明显。Agent可靠性测试中，Hy3工具调用和错误恢复更强，SWE-Bench Verified方差在4%以内。多轮对话任务中，issue率从17.4%降至7.9%，MRCR从42.9%升至75.1%。模型采用Apache 2.0许可，支持vLLM和SGLang部署。
Thank you @ModelScope2022
Thank you @ModelScope2022 ModelScope @ModelScope2022 Meet Hy3 on ModelScope! 295B total / 21B active MoE, built for agentic workflows with 256K context and an FP8 variant ready for deployment. 🚀 License: Apache 2.0. modelscope.ai/collections/Te… BP 🏆 Human eval: 270 experts compared outputs in blind real-world workflow tests. Hy3 scores 2.67/4, ahead of GLM-5.1 at 2.51/4, with the clearest gains in frontend development, CI/CD, and data & storage. 🛠️ Agent reliability: stronger tool calling, formatting, and error recovery, with SWE-Bench Verified variance within 4% across major scaffolds. 🧠 Multi-turn gains: issue rate drops from 17.4% to 7.9%, and MRCR rises from 42.9% to 75.1%. ⚙️ Model lineup: Hy3 BF16 instruct model plus Hy3-FP8 quantized instruct model, with vLLM and SGLang deployment support. 🔗 View Quoted Tweet 💬 1 🔄 2 ❤️ 44 👀 5880 📊 6 ⚡