Arcee AI

9 models across 6 families · Latest: Trinity-Large-Thinking (2026-04)

Researched 19d ago

Agentic AI workflows, efficient and secure

CodingRAGAgentsLong contextVisionClassificationJSON / Tool use

Arcee AI's portfolio covers 9 active models across 6 current families, spanning coding, rag, and agents. Open a model detail page to compare provider routes and sourced benchmarks.
Covers 7 workload areas across 9 active tracked models; last verified 2026-06-29.

Use it for

Teams evaluating coding, rag, and agents across this lab's releases
Comparing model families before committing to a flagship
Migration and pricing follow-ups across 9 tracked models

Do not use it for

Choosing a hosting provider without opening a model page for price ladders

Active models

Current models from this lab, excluding deprecated ones

Active families

Current model families from this lab

Open catalog

9 open

9 open source / 0 open weights

Lowest output price

$0.150 /1M

Cheapest tracked output across active models, per 1M tokens

Latest dated release

2026-04-01

Trinity-Large-Thinking

Freshness

2026-06-29

Researched 19d ago

fresh

Information

Founded2023

San Francisco, California, United States

Links

Website GitHub X / Twitter LinkedIn HuggingFace Crunchbase

Release cadence

Showing 5 recent dated releases (full timeline below). Latest: Trinity-Large-Thinking (2026-04-01).

Where this lab wins

Coding: 1 tracked model with SWE-bench / HumanEval-style scores.
RAG: 6 tracked models with ruler / needle retrieval benchmarks.
Agentic: 7 tracked models with BFCL, tau-bench, and SWE-bench tool-use coverage.
Long-context: 7 tracked models with context-token or InfiniteBench-class signal.

Flagship quality / price signal

Flagship: Trinity-Large-Thinking (best sourced coding quality-per-dollar in this portfolio).

Quality-per-dollar unavailable for this flagship — benchmark coverage or output token pricing is still missing.

Arcee AI is an American AI research organization founded in 2023. Agentic AI workflows, efficient and secure. Arcee AI ships 6 model families totaling 9 models, with the most recent release Trinity-Large-Thinking in 2026-04. Notable families include Trinity, Arcee Maestro, and Arcee Virtuoso. Use it as a stable reference for lab background, release coverage, and follow-up model pages as they are added. View official API endpoints, benchmark performance, and coding/agent fit for every Arcee AI model.

About

Arcee AI, established in 2023, specializes in developing and deploying small language models (SLMs) tailored for business applications. Unlike large language models (LLMs), Arcee AI focuses on creating efficient models designed for specific tasks and industries. Their technology centers around context-adapted and domain-specific LLMs using a system called DALM, along with model merging and advanced post-training techniques. The company offers products like Arcee Conductor, an intelligent model routing solution, and Arcee Orchestra, a platform for designing and scaling AI agents using SLMs. They also provide a model merging tool, MergeKit, and a range of SLMs like Arcee-Blitz and Virtuoso-Large. Arcee AI emphasizes the cost-effectiveness and performance benefits of SLMs, claiming up to 50% cost reduction in training without compromising performance. Arcee AI has raised $29.5 million, including a $24 million Series A round in July 2024. Partnerships with AWS and Arm highlight the potential of their technology. The company's focus on efficient and secure AI solutions positions them well in the evolving AI landscape.

Featured models

Model	Released	Context	Input price ($/1M)	Output price ($/1M)	License	Openness
Trinity-Large-Thinking	2026-04-01	256k	$0.22	$0.85	Apache 2.0	Open source
Trinity-Large-Preview	2026-01-27	128k	$0.15	$0.45	Apache 2.0	Open source
Arcee Maestro Reasoning	2025-12-01	128k	-	-	Apache 2.0	Open source