腾讯混元新模型Hy3，21B激活参数就能比肩更大模型，幻觉率从12.5%降到5.4%，多轮对话能力大增，现在就能在SGLang上跑。
腾讯混元发布Hy3，一个295B参数MoE模型，激活参数仅21B，支持256K上下文。在推理和智能体任务上，Hy3可媲美参数大2-5倍的开源模型。通过反幻觉训练，该模型的幻觉率从12.5%降至5.4%。多轮意图跟踪指标MRCR从42.9%提升至75.1%。Hy3支持MTP+EAGLE推测解码，并提供了FP8检查点以降低部署成本。
🎉 Meet Hy3 from @TencentHunyuan, a 295B MoE with 2…
🎉 Meet Hy3 from @TencentHunyuan, a 295B MoE with 21B active params and 256K context. Day-0 support is now live in SGLang!
1️⃣ Agent-first: rivals flagship open-source models with 2–5x the parameters across reasoning & agentic tasks 2️⃣ Anti-hallucination training: hallucination rate cut from 12.5% to 5.4% in real-world evals 3️⃣ Multi-turn intent tracking: MRCR jumps from 42.9% to 75.1% 4️⃣ MTP + EAGLE speculative decoding out of the box, with an FP8 checkpoint (Hy3-FP8) for cost-effective serving
Cookbook: https://t.co/8IRoSoOLsO
Run it now with SGLang!