Tuomas Rintamaki

Research scientist at Nvidia, working on Nemotron models

[NVIDIA AI @NVIDIAAI]

Today we're shipping Nemotron 3 Ultra.

A 550B MoE frontier-intelligence open model built for long-running agents.

It delivers 5x faster inference and lowers the cost of complex agentic tasks by up to 30% versus other open frontier models.

[vLLM @vllm_project]

🎉 Congrats to @NVIDIAAI on Nemotron 3 Nano Omni — a 30B hybrid Transformer-Mamba MoE (3B active) that unifies vision, audio, video, and text in a single reasoning loop. 256K context, FP8 / NVFP4 quantization, open weights.

Day-0 support in vLLM — tool calling, reasoning, and efficient video sampling for long-video workloads, verified on NVIDIA GPUs.

🔗 https://t.co/95VhKtDLqw
🔗 https://t.co/i8r6VHjVAK

[OpenRouter @OpenRouter]

NVIDIA Nemotron™ 3 Nano Omni is live on OpenRouter.

An open 30B-A3B multimodal model for agentic workflows: text, image, video, and audio in → text out, with a 256k context window and efficient MoE architecture for computer use, documents, and AV reasoning.

[Bryan Catanzaro @ctnzr]

Today we're releasing Nemotron 3 Nano Omni.

Audio, Video, Image, Text ➡️ Text

Ask questions about all your data.

Amazing efficiency powered by the Nemotron Hybrid SSM MoE architecture.

State of the art multimodal intelligence.

[NVIDIA @nvidia]

https://t.co/Ow621ET16t

[Tuomas Rintamaki @tuomars]

Nemotron 3 Nano Omni is out: an open-source Omni model built for strong accuracy, broad capabilities, and speed.
- weights https://t.co/jl3X5OD7E5
- tech report https://t.co/4FY6cEPJBO

[Tuomas Rintamaki @tuomars]

testing posting

[San Francisco Chronicle @sfchronicle]

On Monday, surfers flocked to Mavericks to ride the 30-foot waves.

🎥 courtesy Jack Persons

[@jason @Jason]

Brilliant: drone footage of San Francisco yesterday to blade runner 2049 sound track

[Kyle MacLachlan @Kyle_MacLachlan]

Today is the only day you can RT this. #NationalCoffeeDay

[Josh Patterson @datametrician]

The articles from #GTC19... amazing! @WalmartLabs uses @NVIDIA #GPU and @RAPIDSai to save millions in wasted goods! I'm so proud of the team, and beyond excited for what's coming next. This is just the beginning! https://t.co/UWnuvSXVoO

[Elon Musk @elonmusk]

Oh hi lol

[Fermat's Library @fermatslibrary]

"I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu)”

Linus Torvalds, a 21 year old student at University of Helsinki, announces on Usenet that he is working on a new OS that would eventually become Linux

[The White House 45 Archived @WhiteHouse45]

Earlier this evening, President @realDonaldTrump treated the Clemson Tigers football team to dinner in the State Dining Room! #ALLIN

[Hack Talks @HackTalks]

@tuomars from @NvidiaAI is introducing a brand new topic - Running data science pipelines on the GPU using RAPIDS!

#hacktalks2018 #workshop #AI

[NVIDIA DRIVE @NVIDIADRIVE]

Take a look inside NVIDIA's AI training and validation infrastructure for autonomous driving with MagLev, introduced today at @FB #AtScale #AI #autonomousvehicles https://t.co/Vw1tExZQ2n

[Ian Goodfellow @goodfellow_ian]

GANs for imitating dance moves

[Raymond Hettinger @raymondh]

#Python news: Guido accepted PEP 572. Python now has assignment expressions.

if (match := https://t.co/wHiX17iPKu(data)) is not None:
print(https://t.co/mviw1VeMnU(1))

filtered_data = [y for x in data if (y := f(x)) is not None]

https://t.co/BfCDRi8ZK6

[Clément Farabet @clmt]

Very excited about our work on active learning. For problems where raw data is much larger than what you can effectively label, active learning is key to label the right data. https://t.co/B72hutnlbk