Tencent open-weights Hy4 preview: 770B MoE under Apache 2.0

Tencent released and open-sourced Hy4 preview, its next-generation LLM: a 770B-parameter MoE with 49B active per token and a context window beyond 1M tokens, shipped as standard and FP8 weights under Apache 2.0 (per HF tags; the announcement itself names no license). Target uses are software engineering (planning, debugging, front-end quality), office work (financial and data analysis, cross-document collaboration), game development (playable prototypes from a single prompt) and scientific research. Public benchmark evidence is thin: one internal Tencent blind test with 163 experts over 203 engineering tasks scored Hy4 preview 2.99/4.00, slightly ahead of GLM-5.3 (2.92) and Kimi K3 (2.94). Tencent says the model assisted its own training pipeline and self-optimized its inference stack for a 31.8% end-to-end throughput gain. The full Hy4 is not released yet — a preview-first rollout with the next batch 'soon'. API access via Tencent Cloud TokenHub and OpenRouter at $0.834/M input, $2.501/M output, $0.042/M cached.

This is the third Chinese lab in one week to put major open weights on Hugging Face (Z.ai 8/25, Qwen 8/24-26, Tencent 8/27-28), and at 770B Apache 2.0 it is the closest thing yet to a same-tier, different-organization release for the open-weight frontier — direct evidence for trend #1. But treat capability claims as unverified: no public benchmark scores, one vendor-run blind test, and a preview badge. Serving teams get a vLLM recipe day-0; expect a real read on the model only when public harnesses (SWE-bench, Terminal-Bench) publish runs.