https://zml.ai/topics/serving/ Recent content in Serving on ZML - Model to Metal Hugo en-us Tue, 29 Sep 2026 09:00:00 +0200 https://zml.ai/posts/llmd-20260929.0/ Tue, 29 Sep 2026 09:00:00 +0200 https://zml.ai/posts/llmd-20260929.0/ <p>We&rsquo;re happy to announce the release of <code>zml/llmd 20260929.0</code>, which notably brings <a href="https://hf.co/deepseek-ai/DeepSeek-V4.1-Flash">deepseek-ai/DeepSeek-V4.1-Flash</a> support on NVIDIA and AMD GPUs.</p> <h1 id="models">Models</h1> <p>This release adds 2 model families:</p> <ul> <li><a href="https://hf.co/deepseek-ai/DeepSeek-V4.1-Flash">deepseek-ai/DeepSeek-V4.1-Flash</a> (with native DSpark support)</li> <li><a href="https://hf.co/meta-models/Muse-Glimmer-30B">meta-models/Muse-Glimmer-30B</a></li> </ul> <p>As stated, DeepSeek is only available on platforms with enough VRAM to host it natively. We do not plan to support custom quants in the near future.</p> <h1 id="deepseek-v41-flash">DeepSeek-V4.1-Flash</h1> <p>Obviously the highlight, this is our first release supporting a frontier model. ZML is now mature enough to support complex frontier models, and we wanted to bring it to you. Bear in mind this is a 763B model, weighing 510GB, so for now we only support it on platforms LLMD supports with enough VRAM, namely NVIDIA and AMD GPUs.</p>