NVIDIA releases InternVideo2-CLIP anomaly detection model with TensorRT support
nvidia · hf · 2026-10-23
NVIDIA released nvidia/InternVideo2-CLIP-anomaly-detection on Hugging Face, a model built on InternVideo2-CLIP for video anomaly detection and video-text retrieval, tagged with physical-ai.
Weights ship in TensorRT, ONNX and safetensors formats via the NVIDIA TAO toolchain, with LoRA support for fine-tuning — making it easier to deploy video embeddings and anomaly detection in edge or datacenter pipelines.
Related event: NVIDIA Releases InternVideo2-CLIP Anomaly Detection Model on Hugging Face(2 posts)→
More from Models
- Kimi early tester says seeing a document left him upset and nervous — gnukeith · 2026-10-24
- VraserX: the model isn't the limitation, your prompting is — VraserX · 2026-10-24
- Kimi K3.1 Pro (max) vs Gemini 4 Carbon (high): early tests spark Astra resemblance speculation — lyraxana · 2026-10-24
- Kimi K3.1 vs GPT-6 Astra on one agentic 3D task: half the cost, 16x the reasoning tokens — GapNew4766 · 2026-10-24
- Early Kimi K3.1 Pro access: $4 in tokens, 3-4 hours for Blender clay renders — ChrisGPT · 2026-10-24
- Your model benchmark is meaningless without threshold calibration: 0.5 default missed 48 of 156 — dbreunig · 2026-10-24