Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Today we're releasing Trinity-Large-Thinking. Available now on the Arcee API, with open weights on Hugging Face under Apache 2.0. We built it for developers and enterprises that want models they can inspect, post-train, host, distill, and own.

Arcee.ai

15,607 subscribers

727,389 views • 5 months ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

42 Comments

Arcee.ai5 months ago

Preview showed us where the demand was going. People were already running Trinity-Large-Preview in real agent workflows, with long-horizon tool use and production constraints. So over the last two months, we pushed our SFT and RL stack to meet that moment.

Arcee.ai5 months ago

Our focus was clear. Build a model that stays coherent across turns, uses tools cleanly, follows instructions under constraint, and is efficient enough to serve at scale. That is the bet behind Trinity-Large-Thinking.

Arcee.ai5 months ago

The benchmark picture reflects that bet:

Arcee.ai5 months ago

More important than any one score, Trinity-Large-Thinking is a major step up from Preview in the places that matter most for agents: better multi-turn tool use better context coherence cleaner instruction following more stable long-running behavior

Arcee.ai5 months ago

Read the launch blog for a more detailed breakdown:

Arcee.ai5 months ago

Huge credit to the team that built it, and to the partners and early users who put Preview into real systems and helped shape what came next. Use it today via Arcee OpenRouter > @openclaw is FREE through OpenRouter for the first 5 days! or Hugging Face We are building these models so you can own them.

Arcee.ai5 months ago

Shoutout to @OpenRouter, @digitalocean, @huggingface, @datalogyai, and @PrimeIntellect. Great models need great distribution, great data, and great infrastructure. We are grateful to be building with partners who understand that.

Arcee.ai5 months ago

@digitalocean @huggingface @PrimeIntellect @datologyai - apologies for the typo above!

Florian Brand5 months ago

Congrats!! also, banger joke

Maziyar PANAHI5 months ago

those prices are insanely cheap for a model this size! 👏 claw people are going to be very happy!

Maziyar PANAHI5 months ago

96.3 on AIME25. Beating Opus-4.6 on math. Open weights. This is a genuinely significant release. Congratulations to the entire Arcee team. PS: Anthropic is not having a good week! 😭

John Schulman5 months ago

congrats on the great release!

Sara Hooker5 months ago

Congrats to the Arcee team! 🔥

Petri Kuittinen5 months ago

Trinity-Large-Thinking from Arcee: Apache 2.0 license, a 400B-parameter sparse Mixture-of-Experts with 13B active parameters. 128k/256k context window, 80k max output tokens. Very cheap API pricing: $0.25/M input tokens and $0.90/M output tokens. Text-only. I need to test this!

DigitalOcean5 months ago

Great models deserve great infrastructure. Trinity Large-Thinking is now live in Public Preview on #DigitalOcean. 🌊

Kilo (acq. by Anaconda)5 months ago

Stunning new model. Congrats @arcee_ai team!

Emad5 months ago

Congrats!

Joseph Floyd5 months ago

Congrats squad! Thrilled this is out in the world!

Bojan Jakimovski5 months ago

Wow, this is big !!! Congrats @arcee_ai ♥️

Elaina5 months ago

Another major open model! Congratulations!

Yazan El-Baba5 months ago

Congrats to the team! This is awesome! So glad this isn’t an April Fool’s joke :)

Julian Pscheid5 months ago

Amazing, great work team! Can't wait to try this out in agentic flows. Tagging @togethercompute to add this!

MJ - AI Guy5 months ago

The US needs this. I work with agencies who need elite, trustable (read not Chinese) custom models to run in their higher impact level environments. Awesome work!

No try = no fail (Sean Stromsten)5 months ago

This also looks like fun:

Oleksii Kuchaiev5 months ago

Congratulations on a great model! Always good to see more open models.

Sauers5 months ago

okay

orkward ☄︎5 months ago

apache 2.0 and open weights is the dream. trinity-large is a massive win for devs.

Huepow ↭ ⟠ ⟁∞5 months ago

is their highest model on api, DENSE or loppy MoE? @grok

Hendrix.btc ⭕🔶5 months ago

damn i love the bg sound

Charles Foster5 months ago

Nice!

dan5 months ago

LFG!!!

He-Man, Principal MLE5 months ago

Good, I was getting tired of having to go China for a decent model

岡田泰彦｜光邦｜AI×印刷エヴァンジェリスト5 months ago

This feels like a big shift toward more practical and controllable AI deployment—especially for enterprises. より実用的でコントロール可能なAI活用への大きな一歩に感じます。特に企業にとっては重要ですね。

Ross Taylor5 months ago

Congrats to all involved - very impressive!

Mariusz Kurman5 months ago

Congratulations 🫶🏻

𝕱𝖚𝖑𝖑 𝕶𝖊𝖑𝖑𝖞5 months ago

We're AMERICA and we're gonna with the open source AI race too gawdammit 🫡

WildPinesAI5 months ago

@Scobleizer open reasoning models tuned for agentic tool calling is the gap most labs aren't touching. small team, big swing

Prof Celso Fontes5 months ago

please @arcee_ai consider include this dataset in your next model:

Martin S.5 months ago

free for 5 days is a no-brainer to test openclaw against other agents tbh. ive been running it on openrouter and the throughput is nats

LenaRc !🐀5 months ago

So fucking cool congrats on the release

sam laki5 months ago

lfg

Morgan5 months ago

So awesome - can’t wait to play around with it! 🚀

Related Videos

Today we're open-sourcing nac, an agent harness for long-running tasks, and launching the Arcee open models API beta. Nac is available now under Apache 2.0 on GitHub, built for developers running complex, multi-step engineering workloads.

Today we're open-sourcing nac, an agent harness for long-running tasks, and launching the Arcee open models API beta. Nac is available now under Apache 2.0 on GitHub, built for developers running complex, multi-step engineering workloads.

Arcee.ai

1,695,190 views • 1 month ago

Today, we are introducing Trinity, the start of an open-weight MoE family that businesses and developers can own. Trinity-Mini (26B-A3B) Trinity-Nano-Preview (6B-A1B) Available Today on Huggingface.

Today, we are introducing Trinity, the start of an open-weight MoE family that businesses and developers can own. Trinity-Mini (26B-A3B) Trinity-Nano-Preview (6B-A1B) Available Today on Huggingface.

Lucas Atkins

622,759 views • 9 months ago

Today we're releasing ZONOS2, our next-generation real-time TTS model with high-fidelity voice cloning. ZONOS2 is the most expressive open-source TTS model, released under Apache 2.0 and available on Zyphra Cloud on AMD. 🧵

Today we're releasing ZONOS2, our next-generation real-time TTS model with high-fidelity voice cloning. ZONOS2 is the most expressive open-source TTS model, released under Apache 2.0 and available on Zyphra Cloud on AMD. 🧵

Zyphra

334,157 views • 3 months ago

MagenticLite's models are now fully open source. MagenticBrain and Fara 1.5, previously available on Microsoft Foundry, are live on Hugging Face with open weights. The app, the harness, and every model in the stack are now open.

MagenticLite's models are now fully open source. MagenticBrain and Fara 1.5, previously available on Microsoft Foundry, are live on Hugging Face with open weights. The app, the harness, and every model in the stack are now open.

Microsoft Research

19,398 views • 1 month ago

Open models are how developers build AI they can own. Not just weights. Model families, post-training recipes, evaluation tools, and supporting data give teams the ability to inspect, adapt, and deploy AI for real applications. At GTC Berlin, Bryan Catanzaro, VP of Applied Deep Learning Research at NVIDIA, will show developers what that looks like in practice. 🎙️ Building Open Models That Developers Can Inspect, Adapt, and Deploy: 📍 #NVIDIAGTC Berlin | October 20–22, 2026 🔗 Register with code GTC-NVP8A3H2 for 25% off:

Open models are how developers build AI they can own. Not just weights. Model families, post-training recipes, evaluation tools, and supporting data give teams the ability to inspect, adapt, and deploy AI for real applications. At GTC Berlin, Bryan Catanzaro, VP of Applied Deep Learning Research at NVIDIA, will show developers what that looks like in practice. 🎙️ Building Open Models That Developers Can Inspect, Adapt, and Deploy: 📍 #NVIDIAGTC Berlin | October 20–22, 2026 🔗 Register with code GTC-NVP8A3H2 for 25% off:

NVIDIA GTC

12,309 views • 1 day ago

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.

Harper Carroll

30,392 views • 1 month ago

.River AI has a new vision for personal AI. We co-led River’s $1.1B funding round because we believe in Igor Babuschkin's ambition to build open-weight models that individuals can train and improve continuously. Watch Hemant Taneja in conversation with Igor. Chapters 02:22 — Co-Founding xAI 04:36 — An AI Experience You Own 05:59 — Betting Early on Open Weights 07:47 — How Do You Make Money on Free Models? 09:25 — Why the US Needs Its Own Open Models 11:27 — Enterprises Want to Control Their AI 13:36 — What Is Continual Learning? 16:56 — An API That Updates the Weights 19:17 — A Personal AI for Everyone 20:46 — The 20-Minute Training Run

.River AI has a new vision for personal AI. We co-led River’s $1.1B funding round because we believe in Igor Babuschkin's ambition to build open-weight models that individuals can train and improve continuously. Watch Hemant Taneja in conversation with Igor. Chapters 02:22 — Co-Founding xAI 04:36 — An AI Experience You Own 05:59 — Betting Early on Open Weights 07:47 — How Do You Make Money on Free Models? 09:25 — Why the US Needs Its Own Open Models 11:27 — Enterprises Want to Control Their AI 13:36 — What Is Continual Learning? 16:56 — An API That Updates the Weights 19:17 — A Personal AI for Everyone 20:46 — The 20-Minute Training Run

General Catalyst

112,041 views • 1 month ago

Here's my conversation with Lucas Atkins and the team at Arcee.ai on their path to training and releasing Trinity Large today. From going all in on open models built end to end in the US 6 months ago to having the model in hand is no easy feet. I loved this conversation on how to design a startup around open models and take a bold step to scale it up. I'm openly an Arcee fan, watching them take risk and pull it off. We discuss: - The state (and future) of open vs. closed models, - The business of selling open models for on-prem deployments, - The story of Arcee AI & going “all-in” on this training run, - The ATOM project, - Building frontier model training teams in 6 months, - and other great topics. I really loved this one, and think you well too. Chapters: 00:00:00 Intro: Arcee AI, Trinity Models & Trinity Large 00:08:26 Transitioning a Company to Pre-training 00:13:00 Technical Decisions: Muon and MoE 00:18:41 Scaling and MoE Training Pain 00:23:14 Post-training and RL Strategies 00:28:09 Team Structure and Data Scaling 00:31:31 The Trinity Manifesto: US Open Weights 00:42:31 Specialized Models and Distillation 00:47:12 Infrastructure and Hosting 400B 00:50:53 Open Source as a Business Moat 00:56:31 Predictions: Best Model in 2026 01:02:29 Lightning Round & Conclusions More great open model builder podcasts coming soon!

Here's my conversation with Lucas Atkins and the team at Arcee.ai on their path to training and releasing Trinity Large today. From going all in on open models built end to end in the US 6 months ago to having the model in hand is no easy feet. I loved this conversation on how to design a startup around open models and take a bold step to scale it up. I'm openly an Arcee fan, watching them take risk and pull it off. We discuss: - The state (and future) of open vs. closed models, - The business of selling open models for on-prem deployments, - The story of Arcee AI & going “all-in” on this training run, - The ATOM project, - Building frontier model training teams in 6 months, - and other great topics. I really loved this one, and think you well too. Chapters: 00:00:00 Intro: Arcee AI, Trinity Models & Trinity Large 00:08:26 Transitioning a Company to Pre-training 00:13:00 Technical Decisions: Muon and MoE 00:18:41 Scaling and MoE Training Pain 00:23:14 Post-training and RL Strategies 00:28:09 Team Structure and Data Scaling 00:31:31 The Trinity Manifesto: US Open Weights 00:42:31 Specialized Models and Distillation 00:47:12 Infrastructure and Hosting 400B 00:50:53 Open Source as a Business Moat 00:56:31 Predictions: Best Model in 2026 01:02:29 Lightning Round & Conclusions More great open model builder podcasts coming soon!

Nathan Lambert

26,872 views • 7 months ago

🤝 Arcee.ai, an American AI lab, built its frontier open model family on NVIDIA Blackwell Ultra. Its flagship MoE model, Trinity-Large-Thinking, was RL post-trained using NVIDIA NeMo open libraries. Arcee optimized agentic model inference with NVIDIA Dynamo and vLLM, along with NVIDIA accelerated networking, to deliver low token cost, achieving: ✅ 3T+ tokens served on OpenRouter in its first two months ✅ $0.90 per 1M output tokens ✅ 2nd on PinchBench for open agentic models Explore Arcee's open models, powered by NVIDIA ➡️

🤝 Arcee.ai, an American AI lab, built its frontier open model family on NVIDIA Blackwell Ultra. Its flagship MoE model, Trinity-Large-Thinking, was RL post-trained using NVIDIA NeMo open libraries. Arcee optimized agentic model inference with NVIDIA Dynamo and vLLM, along with NVIDIA accelerated networking, to deliver low token cost, achieving: ✅ 3T+ tokens served on OpenRouter in its first two months ✅ $0.90 per 1M output tokens ✅ 2nd on PinchBench for open agentic models Explore Arcee's open models, powered by NVIDIA ➡️

NVIDIA AI Infrastructure

27,377 views • 2 months ago

Today we're releasing Claude 3 Haiku, the fastest and most affordable model in its intelligence class. Haiku is now available in the API and on for Claude Pro subscribers.

Today we're releasing Claude 3 Haiku, the fastest and most affordable model in its intelligence class. Haiku is now available in the API and on for Claude Pro subscribers.

Anthropic

465,439 views • 2 years ago

Satya Nadella just clarified the Microsoft open Ai deal and the terms what they signed for Azure. Here is exactly how he breaks down the partnership How it was Essentially think of OpenAI having a PaaS business and a SaaS business. The SaaS business is ChatGPT. Their PaaS business is their API. That API is Azure exclusive. The SaaS business, they can run it anywhere. If they want a partner and that partner wants to use a stateless API, then Azure is the place where they can get the stateless API. for even that they'll have to come to Azure. For any custom agreement like that. There are some few exceptions, the US government and so on, that we made, but other than that, they'd have to come to Azure. Open Ai can host there Chat Gpt consumer app wherever they want there wish but when it comes to developers or enterprise want to build on top of Open Ai models Microsoft owned the pipeline.

Satya Nadella just clarified the Microsoft open Ai deal and the terms what they signed for Azure. Here is exactly how he breaks down the partnership How it was Essentially think of OpenAI having a PaaS business and a SaaS business. The SaaS business is ChatGPT. Their PaaS business is their API. That API is Azure exclusive. The SaaS business, they can run it anywhere. If they want a partner and that partner wants to use a stateless API, then Azure is the place where they can get the stateless API. for even that they'll have to come to Azure. For any custom agreement like that. There are some few exceptions, the US government and so on, that we made, but other than that, they'd have to come to Azure. Open Ai can host there Chat Gpt consumer app wherever they want there wish but when it comes to developers or enterprise want to build on top of Open Ai models Microsoft owned the pipeline.

Vikram M

102,664 views • 1 month ago

Today we're announcing Base Labs, a dedicated research organization focused on advancing open-source AI. We believe in a healthy, open frontier model ecosystem. To enable this, we are working on: - Blue-sky research on continual learning, the science of RL, and how models learn across their full lifecycle, with every experiment and recipe shared openly. - The BaseHub Data Foundry: the highest-quality open RL environments, training data, and real-world benchmarks, built for anyone to train and benchmark on. - Post-post training: taking open-source models and making them better, safer, and more aligned through continual post-training, built on our research, and deployed with our frontier safety stack so organizations can use open-source models with confidence. - Making models cheaper and more performant through our model performance research. This is a mission-driven research effort, not a commercial product. We believe the health of the open-source AI ecosystem matters and that the best way to advance it is to do science in the open. We’re hiring engineers, researchers, and research fellows to advance this mission.

Today we're announcing Base Labs, a dedicated research organization focused on advancing open-source AI. We believe in a healthy, open frontier model ecosystem. To enable this, we are working on: - Blue-sky research on continual learning, the science of RL, and how models learn across their full lifecycle, with every experiment and recipe shared openly. - The BaseHub Data Foundry: the highest-quality open RL environments, training data, and real-world benchmarks, built for anyone to train and benchmark on. - Post-post training: taking open-source models and making them better, safer, and more aligned through continual post-training, built on our research, and deployed with our frontier safety stack so organizations can use open-source models with confidence. - Making models cheaper and more performant through our model performance research. This is a mission-driven research effort, not a commercial product. We believe the health of the open-source AI ecosystem matters and that the best way to advance it is to do science in the open. We’re hiring engineers, researchers, and research fellows to advance this mission.

Base Labs

233,795 views • 13 days ago

Today, we’re releasing the open weights for Ling-3.0-flash. 🎉 Official BF16 and FP8-quantized versions are now available, so you can choose the option that best fits your hardware, performance requirements, and deployment needs.

Today, we’re releasing the open weights for Ling-3.0-flash. 🎉 Official BF16 and FP8-quantized versions are now available, so you can choose the option that best fits your hardware, performance requirements, and deployment needs.

Ant Ling

822,868 views • 1 month ago

Introducing Alpamayo 2 Super: the frontier open reasoning VLA for autonomous vehicles and robotaxis. 34B parameters. Full-surround awareness. Available for commercial use under under OpenMDW-1.1. Download now on Hugging Face →

Introducing Alpamayo 2 Super: the frontier open reasoning VLA for autonomous vehicles and robotaxis. 34B parameters. Full-surround awareness. Available for commercial use under under OpenMDW-1.1. Download now on Hugging Face →

NVIDIA DRIVE

17,214 views • 1 month ago

4 years ago we were on the brink of AI becoming proprietary and centralized, when OpenAI kept GPT3 closed and VCs started dumping money on researchers. From fully open science, to fully closed, in a matter of months. It was scary, and 1,000+ leading researchers and scientists banded together to show the world that it was possible to do the same work in the open, and build an ecosystem that benefits everyone. That was the BigScience Research Workshop BLOOM project, and it put us back on track to open science, starting with forward-thinking organizations like Meta releasing OPT. Look at us now. Open models have not only caught up, they're state of the art now. Not just LLMs, but models for document AI, speech to text, text to speech, generating images and more. We're closing in on 2 million open weight models on Hugging Face. Thanks for the reminder Thomas Wolf .

4 years ago we were on the brink of AI becoming proprietary and centralized, when OpenAI kept GPT3 closed and VCs started dumping money on researchers. From fully open science, to fully closed, in a matter of months. It was scary, and 1,000+ leading researchers and scientists banded together to show the world that it was possible to do the same work in the open, and build an ecosystem that benefits everyone. That was the BigScience Research Workshop BLOOM project, and it put us back on track to open science, starting with forward-thinking organizations like Meta releasing OPT. Look at us now. Open models have not only caught up, they're state of the art now. Not just LLMs, but models for document AI, speech to text, text to speech, generating images and more. We're closing in on 2 million open weight models on Hugging Face. Thanks for the reminder Thomas Wolf .

Jeff Boudier 🤗

21,947 views • 1 year ago

🔈 today we're introducing two new live dialogue models Gemini 3.8 Live: built for scale and cost efficiency, combining conversational intelligence with fluid dialogue and visual grounding. Gemini 3.8 Live Extended Thinking: built for high-complexity tasks, with increased intelligence and multi-step reasoning. these audio models are available to build with in AI Studio and via the Gemini API

🔈 today we're introducing two new live dialogue models Gemini 3.8 Live: built for scale and cost efficiency, combining conversational intelligence with fluid dialogue and visual grounding. Gemini 3.8 Live Extended Thinking: built for high-complexity tasks, with increased intelligence and multi-step reasoning. these audio models are available to build with in AI Studio and via the Gemini API

Google AI Studio

63,495 views • 1 day ago

This developer built DeepCrawler - an AI tool that finds API endpoints on any website in under 60 seconds. Built with Next.js 14 + Hyperbrowser SDK. Clean UI, exports to Postman, works on mobile. Useful for developers and security researchers. 100% Open-source

This developer built DeepCrawler - an AI tool that finds API endpoints on any website in under 60 seconds. Built with Next.js 14 + Hyperbrowser SDK. Clean UI, exports to Postman, works on mobile. Useful for developers and security researchers. 100% Open-source

AshutoshShrivastava

78,893 views • 1 year ago

Today, we're open sourcing the code behind "The Infinite Crate," our VST/AU plugin that lets you play with Lyria RealTime directly inside your favorite DAW! 🎧🎶 Since we released The Infinite Crate last year, it’s been used by some of our favorite artists - including a wonderful showcase with Daito Manabe in Tokyo - and being featured as a top new music tool at NAMM 2026. We’re now fully open sourcing the plugin for developers to fork, modify, and make their own under the permissive Apache 2.0 license. 💾 Get it here:

Today, we're open sourcing the code behind "The Infinite Crate," our VST/AU plugin that lets you play with Lyria RealTime directly inside your favorite DAW! 🎧🎶 Since we released The Infinite Crate last year, it’s been used by some of our favorite artists - including a wonderful showcase with Daito Manabe in Tokyo - and being featured as a top new music tool at NAMM 2026. We’re now fully open sourcing the plugin for developers to fork, modify, and make their own under the permissive Apache 2.0 license. 💾 Get it here:

Jesse Engel

20,897 views • 6 months ago