https://www.tanchunsiong.com/tags/glm/ Recent content in GLM on Tan Chun Siong Hugo en-us Tue, 11 Aug 2026 17:00:00 +0800 https://www.tanchunsiong.com/2026/08/updating-openclaw-to-nvidia-glm-5-2-and-minimax-m3/ Tue, 11 Aug 2026 17:00:00 +0800 https://www.tanchunsiong.com/2026/08/updating-openclaw-to-nvidia-glm-5-2-and-minimax-m3/ <p>My earlier OpenClaw setup used NVIDIA DeepSeek V4 Flash with GLM 5.1 as the fallback. Those two models are no longer available on NVIDIA Developer, and NVIDIA started returning an HTTP <code>410</code> response for the old model IDs.</p> <p>I have now replaced them with:</p> <ul> <li>Primary: <code>nvidia/z-ai/glm-5.2</code></li> <li>Fallback: <code>nvidia/minimaxai/minimax-m3</code></li> </ul> <p>The gateway is running with this setup, and I also tested the fallback. GLM timed out, then MiniMax returned <code>OK</code>.</p> <h2 id="the-new-model-setup">The New Model Setup</h2> <p>This is the model order I use in <code>~/.openclaw/openclaw.json</code>:</p> https://www.tanchunsiong.com/2026/08/getting-nvidia-developer-deepseek-flash-working-on-openclaw/ Tue, 11 Aug 2026 16:30:00 +0800 https://www.tanchunsiong.com/2026/08/getting-nvidia-developer-deepseek-flash-working-on-openclaw/ <p>I wanted NVIDIA DeepSeek V4 Flash to be my default model in OpenClaw. I also wanted another NVIDIA model to take over if DeepSeek failed.</p> <p>It took a few changes before tools and follow-up requests worked properly. Here is the setup that worked for me with <strong>OpenClaw 2026.6.11</strong>:</p> <ul> <li>Endpoint: <code>https://integrate.api.nvidia.com/v1</code></li> <li>NVIDIA model: <code>deepseek-ai/deepseek-v4-flash</code></li> <li>OpenClaw reference: <code>nvidia/deepseek-ai/deepseek-v4-flash</code></li> <li>Fallback: <code>nvidia/z-ai/glm-5.1</code></li> </ul> <p>You can find the model details and API examples in NVIDIA&rsquo;s <a href="https://build.nvidia.com/deepseek-ai/deepseek-v4-flash/deploy">DeepSeek V4 Flash API reference</a>.</p>