Vol. I · No. 199WED, NOV 4, 2026
Archive

The Archive

Search the full wire by company, model, lab, or keyword. Every story we have ever aggregated.

Introducing Hy4 Preview

Tencent releases Hy4, a 770B open-weight LLM with 49B active params and 1M token context, 2.6× larger than Hy3.

·

tencent/Hy3

Tencent releases Hy3, a 295B-param MoE model with 21B active params under Apache 2.0, claiming performance parity with 2-5x larger open-source competitors.

·

Tencent Hy 30B/7B/1.8B

Tencent releases Hy-MT2 multilingual translation models (1.8B–30B MoE) supporting 33 languages with 1.25-bit quantization.

··
9 matches