Try our Nemotron 3 Nano Omni Demo at: huggingface.co/spaces/akhaliq…
AK@_akhaliq
Nvidia released Nemotron 3 Nano Omni made a gradio app for it on Hugging Face
Slovenščina
Fuxiao Liu
121 posts
Fuxiao Liu
@FuxiaoL
Research Scientist @Nvidia | CS PhD @UMDCSI, working on LLM, Multimodal Stuff
Washington, DC Katılım Ekim 2021
746 Takip Edilen861 Takipçiler
Fuxiao Liu retweetledi
Stop choosing between efficiency and accuracy for your AI agents. 🛠️ NVIDIA Nemotron 3 Nano Omni is here:
✅ Unified Reasoning: One model for video, audio, and text.
✅ Up To 9x Throughput: Massive efficiency gains for video workflows.
✅ Fully Open: Weights + Data + Recipes. Optimized for Blackwell and available as a NIM. Let’s build. 🚀
Hugging Face: nvda.ws/4u79ue9
Tech Report: nvda.ws/4dbBYxO
#nvidia #nemotron #omni
NVIDIA@nvidia
English
Fuxiao Liu retweetledi
Thank you to everyone in the community who is testing and using Nemotron models. It's great to see Nemotron-Cascade-2, Nemotron-3-Super and Nemotron-3-Nano trending on HF.
The Nemotron team is working hard to incorporate all your feedback into Nemotron 4.
And yes, Nemotron 3 Ultra is still on track for release.
huggingface.co/models?pipelin…
English
What if a VLM could teach itself from zero data?
Meet MM-Zero: one base model that self-evolves by playing 3 roles:
🧠 Proposer – imagines visual concepts
💻 Coder – renders them in Python/SVG
🔎 Solver – reasons over the results
No seed images. Just RL (GRPO) + execution feedback. @zli12321 @ChengsongH31219
AK@_akhaliq
MM-Zero Self-Evolving Multi-Model Vision Language Models From Zero Data paper: huggingface.co/papers/2603.09…
English
🧠 NVIDIA Nemotron 3 Super is the best open-source brain for OpenClaw.
#nvidia #openclaw #clawdbot #nemotron
Bryan Catanzaro@ctnzr
Announcing NVIDIA Nemotron 3 Super! 💚120B-12A Hybrid SSM Latent MoE, designed for Blackwell 💚36 on AAIndex v4 💚up to 2.2X faster than GPT-OSS-120B in FP4 💚Open data, open recipe, open weights Models, Tech report, etc. here: research.nvidia.com/labs/nemotron/… And yes, Ultra is coming!
English
Fuxiao Liu retweetledi
Announcing NVIDIA Nemotron 3 Super!
💚120B-12A Hybrid SSM Latent MoE, designed for Blackwell
💚36 on AAIndex v4
💚up to 2.2X faster than GPT-OSS-120B in FP4
💚Open data, open recipe, open weights
Models, Tech report, etc. here:
research.nvidia.com/labs/nemotron/…
And yes, Ultra is coming!
English
🔥🔥🔥
Bryan Catanzaro@ctnzr
Announcing NVIDIA Nemotron 3 Super! 💚120B-12A Hybrid SSM Latent MoE, designed for Blackwell 💚36 on AAIndex v4 💚up to 2.2X faster than GPT-OSS-120B in FP4 💚Open data, open recipe, open weights Models, Tech report, etc. here: research.nvidia.com/labs/nemotron/… And yes, Ultra is coming!
ART
Fuxiao Liu retweetledi
We remove the requirement of image during the vlm training in MM-Zero!!!
AK@_akhaliq
MM-Zero Self-Evolving Multi-Model Vision Language Models From Zero Data paper: huggingface.co/papers/2603.09…
English
Fuxiao Liu retweetledi
What if a vision model could learn to reason about images without ever seeing one?
MM Zero, from Zongxia Li, @FuxiaoL, and researchers at University of Maryland, Brown, Adobe, and NVIDIA, introduces a three role framework where VLMs bootstrap visual reasoning from literally zero data.
Three agents (Proposer, Coder, Solver) all start from the same base model. The Proposer invents visual concepts and questions. The Coder renders them as SVG code. The Solver reasons over the results. All trained via GRPO reinforcement learning, no human annotation.
On visual reasoning benchmarks, Qwen3 VL 8B improves from 50.7% to 56.6% accuracy. Mimo VL 7B goes from 50.9% to 56.0%. Performance keeps climbing through 5 iterations with no sign of plateau.
The data bottleneck for VLM training may be less about collecting more images and more about letting models generate their own.
English
Today is my last day at xAI, feeling very fortunate about the opportunity. It has been an amazing journey 🫡🫡🫡
English
Same here😭
I haven’t been able to go back since 2019 — the only realistic option is after getting the green card.
Zifan (Sail) Wang@_zifan_wang
I am moving from SF to London! Looking forward to chatting with people here working on AI and AI safety 😏. When I visited my family and friends in China last month, my O1 visa stamp for re-entrance was denied so it is not possible for me to return US immediately after my trip, which was unexpected and crazy. I have been studying and working in US prob like for 7 years. What is the fastest way to adapt to “chips” and drive on the right side 😂?
English
I’ll be attending #neurips2025 in San Diego next week!
If you’re interested in unified models, multimodal agents, or video generation, feel free to reach out.
And if you’re worried that NVIDIA GPUs might not sell out… come talk to me — I promise I can change your mind 😂
I’ll be at the @nvidia booth on Dec 5th #NVIDIA
NVIDIA Newsroom@nvidianewsroom
We’re delighted by Google’s success — they’ve made great advances in AI and we continue to supply to Google. NVIDIA is a generation ahead of the industry — it’s the only platform that runs every AI model and does it everywhere computing is done. NVIDIA offers greater performance, versatility, and fungibility than ASICs, which are designed for specific AI frameworks or functions.
English
Fuxiao Liu retweetledi
We’re delighted by Google’s success — they’ve made great advances in AI and we continue to supply to Google.
NVIDIA is a generation ahead of the industry — it’s the only platform that runs every AI model and does it everywhere computing is done.
NVIDIA offers greater performance, versatility, and fungibility than ASICs, which are designed for specific AI frameworks or functions.
English