Arcee Ai
About Arcee Ai
Arcee AI is a U.S. model lab building foundation-class, open-weight models. Our culture is collaborative: we share experiments early and ship what we learn.
Arcee AI builds open-weight foundation models that run anywhere - on edge, on prem, or cloud. Our Trinity family standardizes capabilities across sizes and is released under Apache-2.0. What you build today runs tomorrow.
Arcee AI is a U.S. model lab led by builders. We began in post-training, then moved upstream to pre-train our own models because we believe progress won’t be defined by size, but by efficiency, adaptability, and the ability to own what you build.
That belief produced Trinity. Our first family of foundation models with consistent skills from on-device to cloud. We optimize for performance per parameter and release weights under Apache-2.0. Every model we ship is permanent and portable, so there’s no lock-in and no forced upgrades. We’re investing to make open weights the durable standard: curated + synthetic data pipelines, reliability measured on tool use and structured outputs, and a commitment to open-weight R&D. The goal is simple: a permanent, practical foundation for AI that is efficient to serve, trustworthy to scale, and ready for real products.
Website ↗Documentation ↗GitHub ↗Hugging Face ↗
Source ↗ · 2026-11-12T02:17:07.907502345Z20 indexed models
20 matching models
arcee-ai/trinity-large-thinking
Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI. It shows strong performance in PinchBench, agentic workloads, and reasoning tasks. Launch video: https://youtu.be/Gc82AXLa0Rg?si=4RLn6WBz33qT--B7...
- Context window
- 262,144
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Artificial Analysis: 10.8
vertexAI/publishers/hf-arcee-ai/models/afm-4.5b@001
- Input modalities
- text
- Output modalities
- text
vertexAI/publishers/hf-arcee-ai/models/afm-4.5b-base@001
- Input modalities
- text
- Output modalities
- text
vertexAI/publishers/hf-arcee-ai/models/arcee-agent@001
- Input modalities
- text
- Output modalities
- text
vertexAI/publishers/hf-arcee-ai/models/arcee-blitz@001
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 40.012%
huggingface/arcee-ai/Arcee-Maestro-7B-Preview
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 23.793%
huggingface/arcee-ai/Arcee-Nova
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 44.053%
azureFoundry/HuggingFace/arcee-ai-arcee-spark@1
`arcee-ai/Arcee-Spark` powered by Text Generation Inference (TGI) - [Example Notebook](https://aka.ms/hugging-face-text-generation-streaming-online-endpoint) - [Original Model Card](https://huggingface.co/arcee-ai/Arcee-Spark) - [Text Generation Inference Documentation](https://
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open Ko-LLM Leaderboard: 70.753%
- Open LLM Leaderboard v2: 28.407%
vertexAI/publishers/hf-arcee-ai/models/arcee-vylinh@001
- Input modalities
- text
- Output modalities
- text
huggingface/arcee-ai/Clown-DPO-Extended
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v1: 76.406%
vertexAI/publishers/hf-arcee-ai/models/llama-3.1-supernova-lite@001
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open Ko-LLM Leaderboard: 52.12%
- Open LLM Leaderboard v2: 30.193%
huggingface/arcee-ai/Llama-Spark
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 27.037%
huggingface/arcee-ai/raspberry-3B
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 15.853%
huggingface/arcee-ai/Saul-Instruct-Clown-7b
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v1: 71.899%
vertexAI/publishers/hf-arcee-ai/models/supernova-medius@001
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open Ko-LLM Leaderboard: 74.777%
- Open LLM Leaderboard v2: 39.154%
vertexAI/publishers/hf-arcee-ai/models/trinity-mini@001
- Input modalities
- text
- Output modalities
- text
huggingface/arcee-ai/Virtuoso-Lite
- Input modalities
- text
- Output modalities
- text
Benchmarks
- Open LLM Leaderboard v2: 36.416%
azureFoundry/HuggingFace/arcee-ai-virtuoso-medium-v2@1
`arcee-ai/Virtuoso-Medium-v2` powered by Text Generation Inference (TGI) - [Example Notebook](https://aka.ms/hugging-face-text-generation-streaming-online-endpoint) - [Original Model Card](https://huggingface.co/arcee-ai/Virtuoso-Medium-v2) - [Text Generation Inference Documenta
- Input modalities
- text
- Output modalities
- text