https://derekwang2002.github.io/ai-intelligence/en/projects/dory/ 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-20261016-02/ ev-20261016-02-published Sat, 17 Oct 2026 15:59:50 GMT New TRIAL <p>NVIDIA released open weights for the 4B-parameter Dory1.0-4B. One checkpoint can run its shared recurrent block one, two or four times, trading more compute for task accuracy. BF16, MXFP8, NVFP4 and Q5_K_M GGUF variants are downloadable with 256K context. Upstream vLLM support remains a pull request, and llama.cpp requires a pinned NVIDIA fork.</p><p><strong>Why read: </strong>One open checkpoint exposes adjustable compute depth worth testing, while stable runtime support is not ready.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16">NVIDIA Dory1.0-4B BF16 model card</a> · <a href="https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-GGUF">NVIDIA Dory1.0-4B GGUF model card</a></p>