Sam Fraser

Trinity-Large-Thinking out today

Check out the blog for a walkthrough on setting up the Hermes Agent powered by Trinity-Large-Thinking across Linux, macOS, and Windows. arcee.ai/blog/how-to-us…

Arcee x @NousResearch Trinity x Hermes Agent

"Trinity-Large-Thinking is noteworthy for the extreme sparsity of its attention mechanism. Its Mixture-of-Experts architecture means that only 1.56% are active for any given token. This allows the model to perform roughly 2-3x faster than its peers on the same hardware. To prevent a few experts from becoming "winners" Arcee developed Soft-clamped Momentum Expert Bias Updates to ensure experts are specialized and routed evenly across a general web corpus."

Switch to ‘dev mode’ for testing of non-default parameters, more advanced settings not available in chat mode, estimated token/credit usage, and various response views. api call updates in real time.

Display the api call for any prompt executed in chat. Copy the call directly into your terminal & update with your key to get started.

We added a few features to our api platform to help users get started with Trinity models:

Arcee’s latest model, Trinity Large Thinking is live now on OpenRouter! It is a 400B total 13B active model with powerful agentic performance, free in @openclaw for the first 5 days!

@mynnoj I worked there for three years before I realized the m365 org was office

super funny that microsoft had a strong brand like 'office' and some mbas decided that 'microsoft 365' and 'copilot' should replace it

Tiny startup Arcee AI built a 400B open source LLM from scratch to best Meta’s Llama techcrunch.com/2026/01/28/tin…

Today, we are releasing our first weights from Trinity-Large, our first frontier-scale model in the Trinity MoE family. American Made. - Trinity-Large-Preview (instruct) - Trinity-Large-Base (pretrain checkpoint) - Trinity-Large-TrueBase (10T pre Instruct data/anneal)

Today, we’re releasing the first weights from Trinity Large, our first frontier-scale model in the Trinity MoE family.

All-in holiday party turned up. Managed to get the McLaren hat without signing up for a crypto account.

Trinity-mini ranks 5th on @xlr8harder's free speach leaderboard speechmap.ai/labs/

We're excited to support @Arcee_ai's Trinity models A family of open base Mixture of Experts models pretrained in collaboration between Arcee, Datology, and Prime Intellect Releasing Trinity Nano (6B) and Mini (26B) today, with Trinity Large still in training

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.

Coming Soon...

Do you understand what this means?

End of day 2 at #allinsummit, such an incredible set of speakers. Still buzzing the @bgurley talk on regulatory capture.