Arcee.ai

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.

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

Huge thanks to @carlfranzen for covering the Trinity-Large-Thinking release for @VentureBeat. We are incredibly proud of what our team accomplished with this release and we cannot wait to see what the open source community builds with it. venturebeat.com/technology/arc…

Jensen showcased PinchBench (by @kilocode) on stage at NVIDIA GTC as the new standard for evaluating @openclaw agent capability. Trinity-Large-Thinking just hit #2 on @pinchbench globally (91.9%) behind only Claude Opus 4.6 (93.3%), which costs ~20x more per token. That's an expensive percentage point... To celebrate the ongoing Trinity partnership, @OpenRouter is running a promotion to make Trinity-Large-Thinking free to use for OpenClaw until Sunday, April 5th. Even when the promo ends, the Trinity economics are pretty incredible. Trinity is $0.25/m input tokens (OpenClaw uses A LOT of input tokens), but with a 60%-90% cache hit rate at $0.06/m cache tokens the average input cost nets out to $0.087/m. You no longer have to compromise on logic to keep your agent infrastructure affordable. We're excited to see what you build.

This April Fool's Day, we decided to stop joking around. Trinity-Large-Thinking is out now.

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

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.

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.

We've seen increased demand for Trinity-Large-Preview, which has resulted in degraded throughput for some users. We're working to restore full capacity. Thank you for your patience, and we'll provide updates when resolved.

hf-mount Attach any Storage Bucket, model or dataset from @huggingface as a local filesystem This is a game changer, as it allows you to attach remote storage that is 100x bigger than your local machine's disk. This is also perfect for Agentic storage!! Read-write for Storage Buckets, read-only for models and datasets. Here's an example with FineWeb-edu (a 5TB slice of the Web): 1️⃣> hf-mount start repo datasets/HuggingFaceFW/fineweb-edu /tmp/fineweb It takes a few seconds to mount, and then: 2️⃣> du -h -d1 /tmp/fineweb 4.1T ./data 1.2T ./sample 5.3T . 🤯😮 Two backends are available: NFS (recommended) and FUSE Let's f**ing go 💪

Here are a few of our favorite shots from our recent out-of-home campaign. Loving how the Arcee teal cuts right through the noise of downtown SF and the traffic on the 101 + a bonus shot from the DC metro.

We've spent a lot of time focused on Trinity Large over the past few months. Meanwhile, Trinity Mini is quietly gaining momentum. Over the last 30 days, official and community derivatives of Mini have crossed 75k downloads on @huggingface. We are also seeing a massive ramp in usage on @OpenRouter throughout March, where it is currently ranked in the top 10 fastest models at 171 tok/s. At 26B, Mini is highly capable and exceptionally fast, making it a great fit for those who need lower latency without sacrificing reasoning. We will continue to support Mini as a core piece of the Trinity family. If you haven't tried it yet, we encourage you to give it a run!