Observed Signal · Jun 2, 2026 · Product Launch · Source: AINews swyx · Impact: 4/5 · Sentiment: Positive
NVIDIA Launches Cosmos 3, Nemotron 3 Ultra, RTX Spark
NVIDIA unveiled multiple AI products including Cosmos 3 — an open, omnimodal family of world models that unifies language, image, video, audio and action — plus Nemotron 3 Ultra, a large MoE open-weight LLM, and the RTX Spark personal AI superchip. Cosmos 3 ships as a full-stack release (weights, code, datasets, fine-tuning recipes) and includes Nano (16B) and Super (64B) model variants, pairing an autoregressive reasoner with a diffusion generator in a Mixture-of-Transformers design. Nemotron 3 Ultra is described as a MoE 550B-A55B open-weight model with community reports of high serving throughput. NVIDIA also previewed RTX Spark (claimed ~1 PFLOP FP4) with Microsoft and other partners, and launched the Cosmos Coalition to foster an open ecosystem for world models. The issue also summarizes contemporaneous multimodal/open-agent releases from MiniMax, Alibaba (Qwen3.7-Plus), JetBrains (Mellum2), and broader trends toward agent runtimes, sandboxes, and local inference tooling.
Major technical releases from NVIDIA (models and hardware) and broad open-model ecosystem activity affect model availability, creative tooling, local inference, and infrastructure choices — all relevant to MarTech/AdTech suppliers and vendors.
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 free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- NVIDIA launched Cosmos 3, an open family of omnimodal world models unifying language, image, video, audio and action in a Mixture-of-Transformers architecture.
- Cosmos 3 model variants include base Nano (16B: 8B reasoner tower + 8B generator tower) and Super (64B: 32B reasoner + 32B generator); Super is finetuned for Text2Image and Image2Video.
- NVIDIA announced Nemotron 3 Ultra, described as a MoE 550B-A55B open-weight LLM; community posts reported high serving throughput (claims of 300+ tokens/sec in some setups).
- NVIDIA previewed the RTX Spark personal AI superchip (claimed ~1 PFLOP FP4, up to 128GB unified memory) with Microsoft and partners (OpenClaw, Hermes Agent) as launch partners.
- NVIDIA released Cosmos 3 as a full-stack open release (weights, code, datasets, fine-tuning recipes) and launched the Cosmos Coalition with partners including Runway to build an open world-model ecosystem.
Connected Companies & Entities
9 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
NVIDIA Nemotron 3 Ultra Went Live June 4
NVIDIA released Nemotron 3 Ultra on June 4, 2026, a 550-billion-parameter hybrid Mamba-Transformer mixture-of-experts (MoE) model with up to ~55B active parameters per token (~90% sparsity) and a 1M-token context window. Trained using NVFP4 (4-bit floating point) on NVIDIA's Blackwell architecture with a hardware-aware 'LatentMoE' expert router, Ultra is delivered as post-trained instruct checkpoints for agent harnesses and is available via build.nvidia.com (NIM microservices), Hugging Face, OpenRouter and select cloud partners. Independent benchmarking (Artificial Analysis) places Ultra at an Intelligence Index of 48 (leading US open-weights releases but trailing some Chinese open models and closed frontier models) and reports inference speeds above 300 tokens/second on a pre-release DeepInfra BF16 endpoint. The article details practical deployment notes: NGC API auth, compute minimums (data-center multi‑GPU/GB200 NVL72), prefer the post-trained instruct checkpoint (not Base), OpenAI-compatible Chat Completions API call patterns, inference defaults (temperature=1.0, top_p=0.95), and cautions about slug lag and vendor claims that require replication.
NVIDIA’s Chris Alexiuk on Nemotron, GPUs, Agentic AI
An interview with NVIDIA’s Chris Alexiuk discussing the Nemotron model family, its evolution, and design choices that align model sizes to GPU hardware. Nemotron 3 ships in three tiers (Nano, Super, Ultra) mapped to single-GPU, single-node, and NVL72 rack deployments. The conversation covers architectural choices (Mamba-2 layers, sparse MoE, reduced attention), LatentMoE routing that compresses tokens to route to more experts, multimodal extensions (Nemotron 3 Nano Omni), and large open data releases alongside model weights. Alexiuk also describes Nemotron 3.5 Lightning as a 30B MoE (3B active) optimized for agent workloads, with speculative decoding and NVFP4 quantization for efficient deployment. He frames NVIDIA’s strategy as enabling broad ecosystem research and production use rather than competing directly as a closed-model vendor.
Nvidia unveils Cosmos 3 world model for robotics
Nvidia announced Cosmos 3, a new 'world model' aimed at robotics and autonomous systems that integrates simulation, scene understanding and action planning into a single foundation model. The article explains what world models are — 3D virtual environments that agents or users can interact with via prompts — and places Nvidia's release alongside DeepMind's Genie 3 (released August 2025), which generates interactive 3D worlds in real time. The piece also discusses research debates about whether large language models possess internal world models, cites reinforcement learning and meta‑learning perspectives (Matthew Botvinick), and describes Yann LeCun's JEPA architecture as an alternative path toward more abstract internal representations. The t3n article was originally published on 2025-08-14 and updated on 2026-06-03 to include Cosmos 3.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.