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(nvidia): correct the GPU fleet claims in the Apps Catalog listing kit
nirholasbf7083f18e60
diff --git a/docs/nvidia-apps-catalog-listing.md b/docs/nvidia-apps-catalog-listing.mdindex c5da2c996..4afe57b44 100644--- a/docs/nvidia-apps-catalog-listing.md+++ b/docs/nvidia-apps-catalog-listing.md@@ -77,10 +77,11 @@ Pick the closest available options in the form; the taxonomy varies.Every item below is live in production unless marked otherwise. Do not claim TensorRT, Triton, or self-hosted NIM microservices; the GPU workers are PyTorch + CUDA today (TensorRT-LLM is an Inception roadmap item).-- **NVIDIA L4 GPUs** (Google Cloud Run GPU) running the entire self-hosted 3D generation fleet: image-to-3D, sketch-to-3D, auto-rigging, text-to-motion, video-to-scene+Also do not claim the **video-to-scene world scanner** or the **texture worker**, and do not re-add **FlashInfer** on their behalf. Both have a GPU `cloudbuild.yaml` in the tree but neither is deployed in any region (verified 2026-08-17 with the recount command below), so they are built code, not a running lane. Deploy them and they belong here; until then a curator who checks would find nothing behind the claim.++- **NVIDIA L4 GPUs** (Google Cloud Run GPU) running the entire self-hosted 3D generation fleet: image-to-3D, sketch-to-3D, auto-rigging, text-to-motion. Eight GPU workers, twelve deployments across `us-central1` and `us-east4` (eleven L4, one RTX PRO 6000). Recount before submitting: `gcloud run services list --project aerial-vehicle-466722-p5 --format="csv[no-heading](metadata.name,region,spec.template.spec.nodeSelector)" | grep -i accelerator`- **CUDA** with custom-compiled extensions (nvdiffrast, diffoctreerast, diff-gaussian-rasterization, torchmcubes)- **NVIDIA Kaolin** (TRELLIS worker)-- **FlashInfer** paged-KV-cache attention (video-to-scene worker)- **NIM APIs** (integrate.api.nvidia.com): Nemotron Super and Nano LLMs, Nemotron Nano VL vision, NV-EmbedQA-E5-v5 embeddings, Mistral 4B reranking, NemoGuard 8B content safety- **NVCF**: microsoft/trellis text-to-3D (the free default forge lane) and FLUX.1-schnell text-to-image- **NVIDIA Riva**: Magpie multilingual TTS and Riva ASR over NVCF gRPCdiff --git a/docs/nvidia-apps-catalog-request.md b/docs/nvidia-apps-catalog-request.mdindex 50819684a..4a67cbd37 100644--- a/docs/nvidia-apps-catalog-request.md+++ b/docs/nvidia-apps-catalog-request.md@@ -16,7 +16,7 @@ three.ws is an NVIDIA Inception member (accepted July 2026). Our product record**Why it belongs in the catalog:** the whole product is NVIDIA accelerated computing end to end, in two layers.-Self-hosted: ten GPU workers on Google Cloud Run NVIDIA L4s, built on `nvidia/cuda` 12.1 to 12.8 images with custom-compiled CUDA extensions (nvdiffrast, diffoctreerast, diff-gaussian-rasterization, torchmcubes) and NVIDIA Kaolin. These run Hunyuan3D 2.1 and TRELLIS for image-to-3D, TripoSG for sketch-to-3D, TripoSR for fast drafts, Make-It-Animatable for auto-rigging, Motion Diffusion Model for text-to-motion, and a video-to-scene world scanner. The heavy image-to-3D lane is validated on RTX PRO 6000 Blackwell with `sm_120` kernel builds.+Self-hosted: eight GPU workers on Google Cloud Run, twelve service deployments across two regions, eleven on NVIDIA L4 and one on RTX PRO 6000 Blackwell. They are built on `nvidia/cuda` 12.1 to 12.8 images with custom-compiled CUDA extensions (nvdiffrast, diffoctreerast, diff-gaussian-rasterization, torchmcubes) and NVIDIA Kaolin, and they run Hunyuan3D 2.1 and TRELLIS for image-to-3D, TripoSG for sketch-to-3D, TripoSR for fast drafts, Make-It-Animatable for auto-rigging, and Motion Diffusion Model for text-to-motion. The heavy image-to-3D lane is validated on RTX PRO 6000 Blackwell with `sm_120` kernel builds.NVIDIA-hosted: Nemotron Super and Nano power agent reasoning, Nemotron Nano VL handles vision, NV-EmbedQA-E5-v5 and Mistral reranking power agent memory, NemoGuard 8B gates every public publish, Riva Magpie TTS and Riva ASR give agents a voice, and Audio2Face-3D drives real-time ARKit blendshape facial animation in the browser (live demo: https://three.ws/demos/audio2face).