Open Source: AI News Week Ending 06/05/2026

Image created with gemini-3.1-flash-image-preview with claude-opus-4.7. Image prompt: A wide 16:9 off-white cover with the words ‘Open Source’ centered and lettered in thick concentric ROYGBIV bands in the style of Julio Le Parc, the ‘O’ of ‘Open’ shaped as an unlatched padlock whose rainbow-banded shackle swings open beside it, flat matte screen-print finish, crisp printed edges, generous negative space, no shadows or gradients.

DeepSeek slated to draw $7 billion in maiden fundraising, sources say
https://www.cnbc.com/2026/06/03/deepseek-slated-to-draw-7-billion-in-maiden-fundraising-sources-say.html

It’s finally here. The official Hermes Desktop app. Available on all platforms.
https://x.com/Teknium/status/2061844602735538266

Introducing the Cosmos Coalition A new global initiative with NVIDIA and leading AI labs to build and open-source frontier world models for physical AI. Runway joins as a founding member, working alongside NVIDIA and a set of leading AI labs to build, share and accelerate world
https://x.com/runwayml/status/2061315089869721682

Jensen just launched NVIDIA Cosmos 3. Pitched as the first fully open omnimodel for physical AI: a mixture-of-transformers (reasoning + generation) with native vision reasoning and generation across text, image, video, sound, and action. Tops open-model leaderboards on
https://x.com/TheHumanoidHub/status/2061333253920080345

NVIDIA Launches Cosmos 3, the Open Frontier Foundation Model for Physical AI | NVIDIA Newsroom
https://nvidianews.nvidia.com/news/nvidia-launches-cosmos-3-the-open-frontier-foundation-model-for-physical-ai

NVIDIA Nemotron 3 Ultra Powers Faster, More Efficient Reasoning for Long-Running Agents | NVIDIA Technical Blog
https://developer.nvidia.com/blog/nvidia-nemotron-3-ultra-powers-faster-more-efficient-reasoning-for-long-running-agents/

We took another look at the capability gap between open-weight and proprietary models. Since the start of the year, open-weight models have lagged the state of the art by four months.
https://x.com/EpochAIResearch/status/2060451576779886942

Verifiers are important for scaling evals/RL But costs add up! So can we make them cheaper? Some great work by @Vtrivedy10 @jakebroekhuizen in conjunction with @nikogrupen @gabepereyra and the Harvey team on this
https://x.com/hwchase17/status/2061867746141356427

👏👏 Introducing Qwen3.7-Plus — a multimodal agent model that unifies vision and language into one versatile agent foundation. ✅ Multimodal interactive hybrid agent: unified GUI & CLI operation across visual and text tasks ✅ Versatile coding agent & productivity assistant with
https://x.com/Alibaba_Qwen/status/2061506641120641494

Qwen
https://qwen.ai/blog?id=qwen3.7-plus

Shared my first trace from @NanoClaw_AI to @huggingface yesterday. Very cool! By default, all agents should store their traces on HF (in private) so that you can keep a history of them, analyze them,… & share them and post-train better models, harnesses and more. Excited
https://x.com/ClementDelangue/status/2062542713463980303

Having trouble connecting Hermes Agent Desktop to your remote instance? Check out this updated guide:
https://x.com/Teknium/status/2062170975949721612

Just pushed an update to help remote connecting with the Hermes Agent GUI over tailscale to function! Please update if you had any issues!
https://x.com/Teknium/status/2061984430370267210

The next evolution of Hermes Agent is here! Introducing Hermes Desktop: everything you love about Hermes, now native on your machine. First demoed in Jensen’s GTC keynote, it’s now in public preview.
https://x.com/NousResearch/status/2061843507417944552

.@GoogleDeepMind’s Gemma 4 – 12B is available on Ollama! Chat: ollama run gemma4:12b-mlx Hermes Agent: ollama launch hermes –model gemma4:12b-mlx Claude Code: ollama launch claude –model gemma4:12b-mlx and more 👇👇👇 (Note, this currently works via MLX)
https://x.com/ollama/status/2062250522598572345

I just shipped oMLX v0.4.0, the first official release with the new native Swift macOS app.
https://t.co/kJeAGjGiSp oMLX now ships with a redesigned onboarding flow, settings UI, Hugging Face cache discovery, and a much more native-feeling way to manage and run local models on
https://x.com/jundotkim/status/2061863850874634242

My new favorite @huggingface feature: hardware compatibility check for every model 🤗 They really are betting on local AI and I love it. Also check out @dphnAI X1 Trinity Nano. 6B params with 3-8-bit versions!
https://x.com/m_newhaus/status/2061824017510584630

Routing and post-training open-source models won’t only give you more accurate systems but also meaningfully faster and cheaper systems as most companies are currently learning (in addition to giving you more control and privacy). The idea that a “”frontier”” model (by frontier we
https://x.com/ClementDelangue/status/2062248714945630632

So much great work lately from Nvidia, the “”King of American Open-source AI””! – Crossed 1,000 total public repositories on @huggingface (820 models, 249 datasets & 57 spaces) & almost 60,000 followers – Current #1 trending model on HF with LocateAnything and #5 trending with PiD
https://x.com/ClementDelangue/status/2061487081315094906

We want to work with kernel developers to help them publish their cool kernels on the @huggingface Hub via🤗 Kernels. This has several advantages: * A consistent build structure * Extreme ease of use * Standardized distribution * Reproducibility Reach out if interested 🤗
https://x.com/RisingSayak/status/2062471134260687264

Automatic behind the scene routing in user interfaces (instead of model picker) will redistribute value capture and usage towards many more models than just frontier ones (especially towards open-source/smaller/cheaper ones). Because it removes the cognitive load for the final
https://x.com/ClementDelangue/status/2061871024627482964

Ideogram 4 not only revamped their model to the best they built yet, but also they flipped from closed to open weights! With a 9B parameter model, I can’t wait to see what the community will do with it Try it now:
https://x.com/multimodalart/status/2062210597148930139

Ideogram 4.0 is now live on fal! SOTA open weights model for realism, text rendering and artistic generation Typography that reads clean on posters, packaging and brand work Lightning fast with multiple quality options for every workflow
https://x.com/fal/status/2062202673361780873

Ideogram just released their latest and best v4 image model open weights State of the art and open weights go well together 🤗 Model:
https://t.co/DUcL7BBH7D Demo:
https://x.com/huggingface/status/2062206083914158287

Introducing Ideogram 4.0: the best open image model in the world. Think it. Make it. Own it. Download the weights, fine-tune on your own data, and run it on your hardware. Live on every Ideogram plan and the API today.
https://x.com/ideogram_ai/status/2062202208700313872

🌞This is big Local AI news! A new open-source Computer-Use LLM has just launched. Holo 3.1 is H Company’s (🇫🇷) new local computer-use agent model that beats Qwen3.5-397B, Kimi-K2.5, and Sonnet 4.6! Since it is built for local deployment → ⬩ Runs fully on your machine
https://x.com/TeksEdge/status/2061825310669332818

Introducing Search Toolkit | Mistral AI
https://mistral.ai/news/search-toolkit/

Major overhaul of the Hermes Dashboard It should now surface a complete management plane. Goal is to reduce or eliminate any needs to run a CLI command directly in your terminal. Let me know if we’re missing anything!
https://x.com/Teknium/status/2062315666439655499

You can use Hermes Desktop with Ollama using local or cloud models. Get started 👇👇👇
https://x.com/ollama/status/2062011585355551231

.@NVIDIA’s Cosmos 3 launched today… and guess who had early access? Agile Robots SE! They’ve been running it across their full portfolio: Thor single- and dual-arm, FR3 Duo. Focus? Simulation. Using Cosmos 3 as a neural simulator; a learned world model that generates
https://x.com/IlirAliu_/status/2061512207738012093

1/ NVIDIA just open-sourced Cosmos 3 at GTC Taipei! It’s the first fully open “”omnimodel”” for physical AI – one model that understands the real world, predicts what happens next, and generates the actions a robot should take. Weights, code, datasets. All open. And this is
https://x.com/kimmonismus/status/2061432501223162241

Breaking news: Cosmos 3 is here. They are attempting to do something completely new 🤯 Why is Physical AI much harder than building a chatbot? Understanding the world is not enough, robots need to predict it and act inside it. That’s the idea behind NVIDIA Cosmos 3: →
https://x.com/TheTuringPost/status/2061308942186414136

In case you missed this: NVIDIA shipped a text-to-image open weights model that looks seriously competitive 👀 (as part of its Cosmos 3 release)
https://x.com/victormustar/status/2061354267546427595?s=20

NVIDIA’s Cosmos 3 is what we’ve never seen before ‒ an Omnimodal World Model. It’s closing the loop for physical AI. All stack in one system: world and multimodal understanding, future generation, reasoning and action This is the next step in Jensen Huang’s AI progression:
https://x.com/TheTuringPost/status/2061474876083274238

NVIDIA’s Cosmos 3 lands at #1 among open weights models in both Text to Image and Image to Video on the Artificial Analysis Leaderboards! Cosmos 3 is a family of omnimodal world models for Physical AI from @nvidia, unifying language, image, video, audio and action in a single
https://x.com/ArtificialAnlys/status/2061494719998546206

is Hermes Agent ready for enterprises? NVIDIA built OpenShell, a runtime that wraps AI agents in the security IT teams need before they let anything touch sensitive systems. it plugs directly into Microsoft’s enterprise security stack. Hermes Agent now runs inside it. what this
https://x.com/shannholmberg/status/2061368566256189656

We have been working closely with @nvidia to ensure Hermes Agent works smoothly on their new @NVIDIARTXSpark superchip and integrates with the new OpenShell runtime, which connects Hermes to @Microsoft’s security primitives. Watch our feature in the big announcement at Computex:
https://x.com/NousResearch/status/2061323987804713083?s=20

@openclaw @NousResearch @LangChain As always, Nemotron 3 Ultra is fully open. This includes model weights, synthetic data, and post-training recipes. Available now on @huggingface →
https://x.com/NVIDIAAI/status/2062521383582646537

🚀 Day-0 support for NVIDIA Nemotron 3 Ultra on vLLM! Ready to be served with the latest vLLM stable release, the new open frontier reasoning model is built for long-running autonomous agents: 🧠 550B total / 55B active — Hybrid Transformer-Mamba MoE 📚 Up to 1M token context
https://x.com/vllm_project/status/2062574262163280172

420.2 tok/s on a 550B model. ⚡️ Nemotron-3-Ultra-550B-A55B reaches 420.2 tok/s powered by BLACKBOX AI Inference Engine. Blackbox now delivers the fastest inference in the industry, outperforming every other provider, including on smaller-parameter models. Check our blog in
https://x.com/blackboxai/status/2062546216949588001

Are you tired of waiting 17 minutes for an AI agent to finish a code change? As an agent’s context grows, standard transformer attention can turn long runs into a bottleneck. @NVIDIAAI Nemotron 3 Ultra addresses this with a hybrid architecture that replaces several
https://x.com/baseten/status/2062609272815685759

Big day for American open models… Nemotron 3 Ultra is now the strongest US open-weight model tested, while apparently serving 300+ tok/s 🤯 Comparable large DeepSeek/Kimi models are usually 50-100 tok/s btw
https://x.com/caspar_br/status/2061505720907182280

Introducing NVIDIA Nemotron 3 Ultra. A frontier smart open model built for long-running agents that need to plan, reason, use tools and keep working across complex coding, research and enterprise workflows. Up to 5x faster inference and up to 30% lower cost for agentic tasks.
https://x.com/nvidia/status/2062522316672667770

Introducing two NVIDIA Nemotron models on Together AI: Nemotron 3 Ultra for high-throughput agentic workloads and Nemotron 3.5 ASR for low-latency multilingual speech recognition. AI natives can now build coding agents, deep research agents, and real-time voice systems on the AI
https://x.com/togethercompute/status/2062520009893576974

nemotron 3 is significantly less sparse than other models (~10% active vs ~3% for kimi K2/deepseek v4)
https://x.com/eliebakouch/status/2061607195268038777

Nemotron 3 Ultra (550B-A55B) is here – our strongest open-weight model and full training recipe to date. Heavy emphasis on real-world inference efficiency for long-context agentic workloads. Everything is open 🤗: base, post-trained, reward checkpoints, NVFP4 quantized
https://x.com/PavloMolchanov/status/2062538679470657727

Nemotron 3 Ultra is now the best open weight model on
https://t.co/EJXiSfWv2O 💚
https://x.com/ctnzr/status/2061483152741175757

Nemotron 3 Ultra was launched today, including a focus on low latency agentic performance. We tested it against peers under restricted turn-usage limits on Terminal-Bench v2.1 – @NVIDIA Nemotron 3 Ultra completes tasks at a much faster pace than peers due to its high inference
https://x.com/ArtificialAnlys/status/2062598349757567359

NVIDIA has just released Nemotron 3 Ultra, the new most intelligent US open weights model, with leading speed for its intelligence Nemotron 3 Ultra scores 47.7 on the Artificial Analysis Intelligence Index, well ahead of the next strongest US open weights models, Gemma 4 31B
https://x.com/ArtificialAnlys/status/2062527871529439438

NVIDIA just announced the release of Nemotron 3 Ultra in Jensen Huang’s Computex keynote: at 550B parameters (55B active), this is the largest Nemotron 3 model to date, and it is the most intelligent US open weights model We partnered with @nvidia to evaluate this model for
https://x.com/ArtificialAnlys/status/2061304911565144230?s=20

NVIDIA just released Nemotron 3 Ultra, a 550b-parameter agentic coding model with a 1m context window. It was built for token efficiency, and is up to 5x faster and 30% cheaper than other similar models. It’s the largest US open-weights model release ever. Free in Cline now!
https://x.com/cline/status/2062620668085297214

NVIDIA Nemotron 3 Ultra is here We have Day‑0 support for Nemotron 3 Ultra in prime-rl and Lab. Specialize Nemotron 3 Ultra for your use case.
https://x.com/PrimeIntellect/status/2062622550300275088

NVIDIA Nemotron 3 Ultra is on Fireworks, day zero. Nemotron Ultra is an open model for frontier reasoning and orchestration in long-running autonomous agents. Think use cases like coding agents, deep research, and complex enterprise workflows. Read on:
https://x.com/FireworksAI_HQ/status/2062568688201646321

NVIDIA’s Nemotron 3 Ultra is available on Ollama’s cloud! Try it 👇 Claude Code: ollama launch claude –model nemotron-3-ultra:cloud Hermes Agent: ollama launch hermes –model nemotron-3-ultra:cloud OpenClaw: ollama launch openclaw –model nemotron-3-ultra:cloud
https://x.com/ollama/status/2062591290743853291

Oh wow, they pre-trained Nemotron 3 Ultra in NVFP4 big update for estimating future model sizes and flops, especially for OpenAI models
https://x.com/scaling01/status/2062540298933219832

The @nvidia Nemotron 3 Ultra is live on CW Serverless Inference 🚀 Open, frontier-reasoning, built for long-running agents — 550B params (55B active), hybrid Transformer-Mamba MoE, up to 1M context. For orchestration, coding agents & deep research. No infra to manage.
https://x.com/wandb/status/2062577626242580896

We are excited to join Nvidia’s Nemotron Coalition of leading AI labs working together to advance open frontier foundation models. To celebrate we have partnered with @nvidia and @nebiustf to provide 2 free weeks of the new Nemotron 3 Ultra model on the Nous Portal!
https://x.com/NousResearch/status/2062554136625766409

With today’s launch of Nemotron 3 Ultra, @nvidia continues to expand its investment in open-source AI. Their flagship frontier-reasoning model, built for long-running autonomous agents, is available Day 0 on Modal. – 550B with 55B active parameters – Hybrid Transformer-Mamba MoE
https://x.com/modal/status/2062528720104227149

Nemotron 3.5 ASR is built for streaming multilingual speech recognition and voice agents. One 0.6B checkpoint. 40 language-locales. Sub-100ms latency. Cache-aware FastConformer carries context forward between chunks instead of repeatedly reprocessing overlapping audio. Try it:
https://x.com/togethercompute/status/2062520605102993436

Open speech models for real-time voice agents. @NVIDIAAI Nemotron 3.5 ASR is now available on fal. An open streaming speech recognition model supporting 40 language-locale combinations with ultra-low latency, native punctuation, and capitalization. Built for voice agents,
https://x.com/fal/status/2062521027020611933

Second big release from us today: Nemotron-3.5-ASR-Streaming! 🌎40 languages ⚡️80ms – 1s controllable latency 🔥240 – 2400 concurrent streams on 1xH100 🧱FastConformer Cache-Aware RNN-T architecture
https://x.com/PiotrZelasko/status/2062538923776290909

It’s been a while! 4 nice additions to the open-weight local-LLM-on-consumer-hardware ecosystem:
https://x.com/rasbt/status/2062235700636873082

Oh wow! Insane open model
https://x.com/cloneofsimo/status/2062210832440918309

Open-weight models have overtaken closed models on OpenRouter. 69.1% of token volume now goes to open-weight models. 30.9% to closed. Competition is a discovery procedure — and developers are discovering the value of open models. 🧵
https://x.com/ttunguz/status/2061846636805177692

Using open models and inference clouds (which serve open models) is a leading indicator of what is to come. The advantage of open weights is that you can train, serve, and continually improve your own model. You get control over cost, quality, latency, deployment options, etc.
https://x.com/ypatil125/status/2062196581936529721

We 💚 open source
https://x.com/NVIDIAAI/status/2061498958283968735

langsmith! ✅ Sandbox:
https://t.co/vaChlwHbrm ✅ Gateway:
https://t.co/UqWDeBFS2H ✅ Observability:
https://x.com/hwchase17/status/2062144718427857256

VoidZero, the team behind Vite, Vitest, Rolldown, Oxc, and Vite+, is joining Cloudflare. Vite stays open source, vendor-agnostic, and built for everyone.
https://x.com/Cloudflare/status/2062521221132992533

Congrats to the @Alibaba_Qwen team, the new Qwen-3.7 Plus is a best in class multimodal model, performing near SOTA as both a coding and general purpose agent across benchmarks. Free to try in Cline now! `npm i -g cline`
https://x.com/cline/status/2061580233778790439

[2606.03746] Qwen-Image-Flash: Beyond Objective Design
https://arxiv.org/abs/2606.03746

A practical step forward for real-world manipulation: an open-source world model that replaces rigid action chunking with event-grounded prediction. It anchors planning and control to actual physical moments (reach → grasp → contact → place), giving robots more natural
https://x.com/IlirAliu_/status/2060422419479974312

The first open-source unified world model for scalable robot manipulation: 5B-parameter open-source unified video-action world model that combines policy and world modeling to generate robot actions, predict future visuals, and evaluate task progress from observations, language,
https://x.com/IlirAliu_/status/2061870964519076312

Open and closed models are on different exponentials
https://www.interconnects.ai/p/open-and-closed-models-are-on-different