Tencent (Hunyuan) · MoE Multimodal LLM
Hy4-flash
Hy4-flash is a mixture-of-experts model with 220B total parameters (including 64B Engram parameters) and 12B active per token, accepting text and image inputs with a 1M-token context window. It adds hybrid sparse and sliding-window attention, LatentMoE routing and n-gram embeddings, plus a built-in 0.85B speculative drafter (DFly2). The page compares its base model with Hy4 Base and DeepSeek-V4-flash-Base on knowledge, reasoning, math, code and multimodal benchmarks. It is released under the Tencent Hy licence on Hugging Face, ModelScope and GitCode.
Open the source ↗Similar Language Models releases
- ERNIE 5.5 Flash Preview
Baidu · Agentic MoE coding model · Week 47, 2026
- ERNIE 5.5 Flash Base
Baidu (ERNIE) · Open MoE base model · Week 46, 2026
- Hy4 preview
Tencent Hy Team · Open MoE LLM · Week 46, 2026
- ERNIE 5.5 Flash Base
Baidu · MoE Base Model · Week 46, 2026
- Clover-35B-A3B-Preview
DiffuSpace · Diffusion MoE LLM · Week 45, 2026
- MemT-1.0-Nano
MemTensor · Memory LLM · Week 44, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.