Tencent releases Hy4 flash
Cheaper Hy4 tier: 224B parameters with 12B active, multimodal with a 1M-token context; weights planned.
We’re introducing Hy4 flash, our most cost-effective Hy model to date—featuring 224B total parameters (including 64B Engram parameters) with just 12B active parameters per token.
Official source
Introducing Hy4 flash: Ultimate cost-efficiency for everyday productivityhy.tencent.ai · published
Follow-ups
Model
Hy4 flash
Tencent · Announced
- DeepSWE v1.164.6%
- HLEWith tools, full set (2,500 questions)55.2%
- Terminal-Bench 4.019.2%