WhynotGPT/A100-Dinov3-H-SFT

AI model by WhynotGPT

Older snapshot: check the original source for current information.

From the publisher

Model card & documentation

Source preview

Read the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.

30,000-step SFT checkpoint from sftv1dinov320261029r2. Code and complete instructions: mirrorbloom/EgoNet · A100-Infer. The download script pins the weight revision and checks SHA256. Full-frame RGB, optionally calibrated pinhole K and aligned optical-axis Z-depth. Supports RGB, RGB+K and RGB+K+Depth feature modes. Known K is retained for geometry even in RGB feature mode; missing K uses the registered preset. Brown-distorted ARCTIC RGB follows training-time rectification. Native fisheye is outside this first SFT version. Offline bidirectional T8, aspect-preserving resize/pad to 512×896, no hand crop. Outputs two hands in fixed left, right order: camera-space meters, 21 joints, 778 vertices, full MANO rotations and shape, wrist/middle-MCP positions, plus trained presence scores. Short clips use invalid temporal padding. The exported checkpoint preserves all 900 model tensors unchanged, including DINOv3 H+, projection, MoT, temporal modules, all task heads, presence and MANO buffers. Only optimizer/RNG are omitted. Original DINO initialization weights and external MANO pkl files are unnecessary for…

WhynotGPT/A100-Dinov3-H-SFT on Hugging Face A short preview, not the full document.

Read the full model card ↗ · Preview checked 2026-10-29T09:30:56.458Z

Inside the original model card — Document outline

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

Model overview

WhynotGPT/A100-Dinov3-H-SFT is a task-unspecified model repository published by WhynotGPT on Hugging Face. The source reports the pytorch library.

This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.

Model facts

Task: Not reported
Library: pytorch
Recent downloads (30 days): 0
Cumulative likes: 0
Hugging Face trending score: 0
Reported safetensors parameters: Not reported
Architecture: Not reported
License: Not reported — check the model card
Access: Not gated by Hugging Face
Created: 2026-10-29T06:13:14.000Z
Last modified: 2026-10-29T06:24:41.000Z

Compatibility and lineage

Language coverage was not reported.

Reported base models: facebook/dinov3-vith16plus-pretrain-lvd1689m

Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.

Use and evaluate

Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.

No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.

Model card and usage instructions

Model files and configuration

Community discussion

Source and freshness

Source: Hugging Face Hub. Metadata observed 2026-10-29T06:48:44.800Z. Daily imports are snapshots, not real-time monitoring.

Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.

Open original source