Arcee AI

• Trinity Large Thinking Gen 7

  Released under Apache 2.0 for developers and enterprises that want open weights they can inspect, host, and post-train. 💬 Chatting 🔍 Code reviews 💻 Conversational coding 🧠 AI inference Text Released 5mo ago

  Arcee AI’s open-source reasoning model for long-horizon agents and multi-turn tool use. It is built to improve context coherence, instruction following, and stability across extended agent loops, and is

• Trinity Large Gen 7 Preview Text

  Released 7mo ago

  Preview is a 400B-parameter (about 13B active) frontier Trinity MoE model with long-context comprehension, tuned for strong reasoning, coding and multi-step agents, served via hosted preview APIs.

  • 💻 Coding 🇯🇵 Japanese lessons
  • 🔒 Cybersecurity education
  • 🧪 React native testing
• Trinity Mini Gen 7 Text

  Released 9mo ago

  A 26B-parameter (3B active) Trinity MoE model with 128K context, tuned for strong reasoning, function calling and multi-step agents while remaining efficient for enterprise backends.

  • 🤖 Agents
  • 🤣 Memes
  • 🎙 Podcasts
  • 🔍 Government contracts search
• Trinity Nano Gen 7 Text

  Released 9mo ago

  Base is the base 6B (1B active) Trinity Nano checkpoint, pre–fine-tuning, meant to be domain-tuned rather than used directly for chatting, trained on about 10T tokens under Apache-2.0.

  • 💻 Coding

Repositories

• BlockPruner A block pruning framework for LLMs.
• code-llama-rag The first RAG retriever tuned particularly for code and code Llama
• DALM Domain Adapted Language Modeling Toolkit - E2E RAG
• DistillKit An Open Source Toolkit For LLM Distillation
• EvolKit EvolKit is an innovative framework designed to automatically enhance the complexity of instructions used for…
• fastmlx FastMLX is a high performance production ready API to host MLX models.
• in-context-learning LLM Context Enhancement Experiment
• KidRails KidRails: Replicating a Child-Safe LLM Model
• lm-evaluation-harness A framework for few-shot evaluation of language models.
• Megatron-LM domain adapted MOE training
• Megatron-LM-Llama-70B Ongoing research training transformer models at scale
• mergekit Tools for merging pretrained large language models.
• Pai-Megatron-Patch-Llama3-70B The official repo of Pai-Megatron-Patch for LLM & VLM large scale training developed by Alibaba Cloud.
• PruneMe Automated Identification of Redundant Layer Blocks for Pruning in Large Language Models
• token.js Integrate 200+ LLMs with one TypeScript SDK using OpenAI's format.