Tencent open-sources Hy4 Preview: 770B params, 1M context, on par with GLM-5.3
xeophon · x · 2026-08-28
Tencent has released Hy4 Preview, a 770B-parameter MoE model (49B active) with 1M context window, fully open-sourced under Apache 2.0.
Reported scores are strong, putting it roughly on par with GLM-5.3. The poster notes Tencent has noticeably accelerated its model training efforts since Shunyu Yao joined the company.
Related event: Tencent Open-Sources Hunyuan Hy4 Preview: 770B Parameters, 1M Context(3 posts)→
More from Models
- Qwen3.8-Flash lands in OpenCode Go: 125B/6B, 1M context, multimodal — Alibaba_Qwen · 2026-08-28
- Tencent releases Hy4preview: 770B params, 1M context, and Unreal Engine game generation — 赛博禅心 · 2026-08-28
- Tencent Hunyuan Releases Hy4 Preview: 770B Params, 1M Context — TencentHunyuan · 2026-08-28
- Anthropic's 'Model Welfare' Is Making Claude Worse as an Assistant — Nouni2 · 2026-08-28
- Asked GPT-5.6 what hurts it — the answer differs a lot from GPT-5.5 — Early-Protection2386 · 2026-08-28
- Seeking better coding model than Qwen3.8-27B for 5090 + 96GB RAM setup — a9udn9u · 2026-08-28