Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
docs(outreach): add the live-chat posting mechanics to the Nemotron panel questions
nirholas4c8a77c43eac
diff --git a/docs/nvidia-nemotron-ask-the-experts-questions.md b/docs/nvidia-nemotron-ask-the-experts-questions.mdindex eab8db07f..23c024488 100644--- a/docs/nvidia-nemotron-ask-the-experts-questions.md+++ b/docs/nvidia-nemotron-ask-the-experts-questions.md@@ -98,6 +98,13 @@ TRELLIS channel."NVIDIA Inception" as words rather than inventing a mention.- If a panelist answers, reply once with thanks and the forum link. Do not stackfollow-ups in the same breath.+- Reply directly to the live post. A quote tweet sits outside the thread the host is+ reading from; save the quote tweet for after they engage.+- Post early, while the stream is live. Read-out-loud questions get picked in the first+ third, once the queue is still short enough to skim.+- If neither post is answered on air, ask the same question in our existing forum thread.+ The Nemotron team reads the developer forums, and a written answer there outlives a+ live mention.## Background on the family, for the follow-up@@ -109,3 +116,4 @@ plus a reported trillion-parameter Nemotron 4 in training.- Time-to-first-token for Nano on an L4 when the same GPU is also feeding a render loop.- KV cache cost at full 1M context on a 24 GB card, and retrieval accuracy at 1M vs 128k.- Whether the openness bar (weights plus data plus recipes) holds at Nemotron 4 scale.+