https://derekwang2002.github.io/ai-intelligence/en/projects/hy4/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT https://derekwang2002.github.io/ai-intelligence/en/events/ev-20260828-02/ ev-20260828-02-adoption-review-20260927 Sun, 27 Sep 2026 16:01:12 GMT Update WATCH <p>Hy4 preview reached 22,788 monthly downloads, but the full model and reproducible independent evaluation remain absent.</p><p>Original story: Tencent open-weights Hy4 preview: 770B MoE under Apache 2.0</p><p><strong>Why read: </strong>Downloads grew sharply, but independent reproduction remains the key trial gate.</p><p>Recommendation: WATCH</p><p>Sources: <a href="https://huggingface.co/tencent/Hy4-preview">huggingface.co · Hy4 preview</a> · <a href="https://www.tencent.com/tencent-releases-and-open-sources-tencent-hy4-preview">tencent.com · tencent releases and open sources tencent hy4 preview</a></p> https://derekwang2002.github.io/ai-intelligence/en/events/ev-20260828-02/ ev-20260828-02-published Sun, 30 Aug 2026 00:00:20 GMT New WATCH <p>Tencent released Hy4 preview as a 770B-parameter MoE with 49B active parameters per token and a context window beyond 1M tokens. Standard and FP8 weights use Apache 2.0. Tencent&apos;s announcement reports an internal expert blind test and a 31.8% inference-throughput gain from model-assisted optimization. The Hugging Face page now exposes public leaderboard metadata, including SWE-bench Pro and Terminal-Bench 2.1 scores, but it does not identify a reproducible run or an independent evaluator. The full Hy4 is still absent; only the preview and FP8 repositories are listed.</p><p>Recommendation: WATCH</p><p>Sources: <a href="https://www.tencent.com/tencent-releases-and-open-sources-tencent-hy4-preview">tencent.com · tencent releases and open sources tencent hy4 preview</a></p>