Observed Signal · Sep 10, 2026 · Technical Release · Source: https://martechseries.com/feed/ · Impact: 3/5 · Sentiment: Positive
Versos AI Launches NVIDIA NeMo Agents for Video Training Data Curation
Versos AI, a provider of video training data infrastructure for library owners and AI labs, announced new capabilities built with NVIDIA NeMo that let AI teams describe dataset needs in natural language, enabling search, evaluation, and assembly of rights-cleared video into structured training datasets. The platform, now listed in AWS Marketplace, connects to Amazon S3 to prepare, manage, and monetize video archives for AI model training without data migration, creating metadata, embeddings, and provenance. It supports private offers and AWS spending commitments. Built on NVIDIA CUDA, Nemotron Ultra, and LangChain, the system accelerates model development and enhances video library discoverability. These advancements follow Tape Ark partnership and NVIDIA Inception acceptance. Versos AI will demonstrate at IBC2026 in Amsterdam.
New Agentic AI capability for video training data curation addresses a critical bottleneck in AI model development for content owners and AI labs. Could facilitate licensing and monetization of video libraries, impacting the content and advertising ecosystem.
Track NVIDIA Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Versos AI launched new NVIDIA NeMo-based natural language video dataset assembly.
- Platform now available in AWS Marketplace, integrating with Amazon S3.
- Supports private offers and AWS spending commitments with no data migration.
- Built on NVIDIA CUDA, Nemotron Ultra, and LangChain.
- Will be demonstrated at IBC2026 in Amsterdam, September 11-14, 2026.
Connected Companies & Entities
3 Entities mapped“The workflow uses NVIDIA CUDA Toolkit, NVIDIA Nemotron Ultra, and LangChain.”
“LangChain provides the agent scaffolding layer that coordinates the steps between a plain-language request and the platform's functions.”
“AWS Marketplace helps organizations discover, buy, and manage software solutions.”
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
VIDIZMO joins NVIDIA Inception for agentic AI
VIDIZMO Inc., an enterprise AI orchestration and applications platform, has joined NVIDIA Inception, a program supporting companies building on the NVIDIA platform. The membership provides VIDIZMO engineering teams access to NVIDIA developer tools, technical training, preferred pricing, cloud credits, and co-marketing support. VIDIZMO specializes in on-premises and air-gapped AI deployments, ensuring customer data remains within their control. The company's products, including VIDIZMO Nexus, AI Intelligence Hub, AI Live Insight, and Redactor, run on NVIDIA technologies such as DeepStream SDK, TensorRT, and NIM microservices. This partnership aims to accelerate VIDIZMO's AI capabilities and support its customers in regulated industries like public safety, government, healthcare, and financial services.
Novi AI Launches Long Video Agent
Novi AI announced the launch of its Long Video Agent, an end-to-end AI workflow that converts a story idea, script, or screenplay into a complete narrative video up to five minutes long. The system consolidates script generation, multi-scene visual production, synchronized voiceover, and editing into a single process and is aimed at creators producing AI animated series, short dramas, and episodic content. Novi AI says its platform serves over 2 million creators worldwide and the Long Video Agent integrates multiple generation models (Seedance 2.0, PixVerse C1, Wan 2.7). The product is available now on Novi AI’s platform with ongoing updates planned.
NVIDIA’s Nemotron 3 Nano Omni Multimodal Model
NVIDIA published a research paper introducing Nemotron 3 Nano Omni, a single unified multimodal model that natively ingests and reasons across text, images, video and audio. The model uses a Mixture-of-Experts (MoE) backbone (described as a 30B total / ~3B active configuration), vision and audio encoders named C-RADIOv4-H and Parakeet-TDT, dynamic-resolution image handling, Conv3D-based temporal compression and Efficient Video Sampling for video. Nemotron 3 increases working memory to 256,000 tokens and ships quantized variants (BF16, FP8, FP4) intended to enable inference on more modest hardware. The paper reports substantial throughput and per‑GPU efficiency gains versus competitors and provides model weights and training details via an arXiv preprint.
Receive all future signals for NVIDIA & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.