# NVIDIA: what I've logged

> NVIDIA comes up in 12 posts of my AI log, from December 2025 to October 2026, and each item below links to the primary source. The releases I logged include Nemotron 3.5 Lightning, Nemotron 3 Ultra and GTC Taipei at Computex 2026, filed under frontier, open weights and robotics.

## Items

1. **NVIDIA Run:ai Is Now Open Source as Part of NVIDIA DSX AI Factory Platform** (NVIDIA, 13 Oct 2026): The full Run:ai GPU orchestration platform is open source, with KAI Scheduler and a GPU Fractioning Operator usable on their own. [Primary source](https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform) · [Post](https://heyaliasif.com/ai/2026-10-14-gemini-4-argon-lfm-runai#item-4)
2. **Nvidia VFX Nodes for Video Upscaling, Frame Generation and SDR-to-HDR** (ComfyUI, 13 Oct 2026): RTX super resolution, frame generation and SDR-to-HDR now run as local ComfyUI nodes that can be chained in one workflow. [Primary source](https://blog.comfy.org/p/nvidia-vfx-nodes-for-video-upscaling) · [Post](https://heyaliasif.com/ai/2026-10-14-gemini-4-argon-lfm-runai#item-5)
3. **ReplaySSM: Rethinking State Storage for Faster Hybrid-Model Decoding** (NVIDIA, 12 Oct 2026): Stores state-space layer state as a checkpoint plus recent inputs, cutting memory traffic and simplifying speculative decoding rollback. [Primary source](https://developer.nvidia.com/blog/replayssm-rethinking-state-storage-for-faster-hybrid-model-decoding) · [Post](https://heyaliasif.com/ai/2026-10-13-muse-wonder-self-replication-radar#item-6)
4. **Nemotron 3.5 Lightning** (NVIDIA, 10 Aug 2026): A 30B hybrid Mamba-Transformer with only 3B active parameters and up to 1M-token context, open weights in several precisions. [Primary source](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16) · [Post](https://heyaliasif.com/ai/2026-08-10-week-in-ai#item-5)
5. **Nemotron 3 Ultra** (NVIDIA, 1 Jun 2026): A 550B mixture-of-experts with 55B active, a 1M-token context and NVFP4 pretraining, released June 4 with NVIDIA claiming 5.9 times the throughput of GLM-5.1. [Primary source](https://research.nvidia.com/labs/nemotron/Nemotron-3-Ultra/) · [Post](https://heyaliasif.com/ai/2026-06-01-week-in-ai#item-1)
6. **GTC Taipei at Computex 2026** (NVIDIA, 1 Jun 2026): GR00T 1.7 in early access, Isaac Sim 6.0 generally available and the first open humanoid reference design. [Primary source](https://blogs.nvidia.com/blog/nvidia-gtc-taipei-computex-2026-news/) · [Post](https://heyaliasif.com/ai/2026-06-01-week-in-ai#item-5)
7. **Cosmos 3: Omnimodal World Models for Physical AI** (arXiv, 1 Jun 2026): The technical report behind NVIDIA's new world model. [Primary source](https://arxiv.org/abs/2606.02800) · [Post](https://heyaliasif.com/ai/2026-06-01-week-in-ai#item-7)
8. **Cosmos 3** (NVIDIA, 25 May 2026): An open omni-model that combines world generation, physical reasoning and action in one mixture-of-transformers, announced at Sunday's GTC Taipei keynote. [Primary source](https://huggingface.co/blog/nvidia/cosmos-3-for-physical-ai) · [Post](https://heyaliasif.com/ai/2026-05-25-week-in-ai#item-2)
9. **Nemotron 3 Nano Omni** (NVIDIA, 27 Apr 2026): An open 30B-A3B model that takes video, audio, image and text and acts as the perception layer for agent systems. [Primary source](https://blogs.nvidia.com/blog/nemotron-3-nano-omni-multimodal-ai-agents/) · [Post](https://heyaliasif.com/ai/2026-04-27-week-in-ai#item-4)
10. **NemoClaw** (NVIDIA, 16 Mar 2026): Nvidia packaged security and governance controls around OpenClaw-style agents so companies can run them with managed inference. [Primary source](https://nvidianews.nvidia.com/news/nvidia-announces-nemoclaw) · [Post](https://heyaliasif.com/ai/2026-03-16-week-in-ai#item-4)
11. **Nemotron 3 Super** (NVIDIA, 9 Mar 2026): A 120B mixture-of-experts model with 12B active parameters and a 1M-token context, released with weights and training recipe. [Primary source](https://nvidianews.nvidia.com/news/nvidia-announces-nemotron-3-super) · [Post](https://heyaliasif.com/ai/2026-03-09-week-in-ai#item-1)
12. **Alpamayo** (NVIDIA, 5 Jan 2026): Nvidia opened a family of reasoning models, tools and data for autonomous driving, built around rare edge cases. [Primary source](https://nvidianews.nvidia.com/news/nvidia-announces-alpamayo-family-of-open-source-ai-models-and-tools-to-accelerate-safe-reasoning-based-autonomous-vehicle-development) · [Post](https://heyaliasif.com/ai/2026-01-05-week-in-ai#item-4)
13. **Vera Rubin platform** (NVIDIA, 5 Jan 2026): Nvidia said its next six-chip platform is in full production, with a claimed large cut in cost per token against Blackwell. [Primary source](https://nvidianews.nvidia.com/news/rubin-platform-ai-supercomputer) · [Post](https://heyaliasif.com/ai/2026-01-05-week-in-ai#item-6)
14. **NVIDIA and Groq non-exclusive inference licensing agreement** (Groq, 22 Dec 2025): The biggest chip company licensing a fast-inference specialist's technology says inference speed is now the contested ground. [Primary source](https://groq.com/newsroom/groq-and-nvidia-enter-non-exclusive-inference-technology-licensing-agreement-to-accelerate-ai-inference-at-global-scale) · [Post](https://heyaliasif.com/ai/2025-12-22-week-in-ai#item-4)
15. **NVIDIA Nemotron 3 Nano 30B-A3B** (NVIDIA, 15 Dec 2025): An open mixture-of-experts model with only about 3B active parameters, aimed at cheap multi-agent workloads. [Primary source](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16) · [Post](https://heyaliasif.com/ai/2025-12-15-week-in-ai#item-3)
16. **Alpamayo-R1** (NVIDIA, 1 Dec 2025): An open 10B vision-language-action model for autonomous driving that explains its decisions. [Primary source](https://huggingface.co/nvidia/Alpamayo-R1-10B) · [Post](https://heyaliasif.com/ai/2025-12-01-week-in-ai#item-5)

## Posts

- 2026-10-14 [Gemini 4 Argon lands, Run:ai goes open source](https://heyaliasif.com/ai/2026-10-14-gemini-4-argon-lfm-runai)
- 2026-10-13 [Muse Wonder ships without scores, Anthropic tests self-copying](https://heyaliasif.com/ai/2026-10-13-muse-wonder-self-replication-radar)
- 2026-08-10 [Grok 4.6, Gemini 3.7 Flash and GLM-5.3 arrive in four days](https://heyaliasif.com/ai/2026-08-10-week-in-ai)
- 2026-06-01 [Nemotron 3 Ultra ships, MiniMax M3, Gemma 4 12B, Glasswing grows](https://heyaliasif.com/ai/2026-06-01-week-in-ai)
- 2026-05-25 [Claude Opus 4.8, NVIDIA's Cosmos 3 keynote, Codex on Windows](https://heyaliasif.com/ai/2026-05-25-week-in-ai)
- 2026-04-27 [Grok 4.3, Mistral Medium 3.5 weights, and the Qwen3.6 family](https://heyaliasif.com/ai/2026-04-27-week-in-ai)
- 2026-03-16 [GTC week: NemoClaw, GPT-5.4 mini and nano, and Mistral Small 4](https://heyaliasif.com/ai/2026-03-16-week-in-ai)
- 2026-03-09 [Nemotron 3 Super opens up and Gemini Embedding 2 goes multimodal](https://heyaliasif.com/ai/2026-03-09-week-in-ai)
- 2026-01-05 [CES week: Nvidia's open driving model, Atlas for factories, LTX-2](https://heyaliasif.com/ai/2026-01-05-week-in-ai)
- 2025-12-22 [GLM-4.7 and MiniMax M2.1 open up, Nvidia licenses Groq's tech](https://heyaliasif.com/ai/2025-12-22-week-in-ai)
- 2025-12-15 [Gemini 3 Flash, GPT-5.2-Codex and Nvidia's open Nemotron 3](https://heyaliasif.com/ai/2025-12-15-week-in-ai)
- 2025-12-01 [DeepSeek V3.2, Mistral 3, and Anthropic buys Bun](https://heyaliasif.com/ai/2025-12-01-week-in-ai)

Canonical: https://heyaliasif.com/ai/topics/nvidia