# lean-think-heavy - leanmodels.ai

*Heavyweight class, v1.0, Arcee Trinity Large-Thinking base*

Extended reasoning - chain-of-thought, agentic tasks, deep analysis

398B total, ~13B active per token. Interleaved sliding-window + global attention with chain-of-thought and agentic RL post-training. The largest model in the catalog.

## Specifications

- **Total params:** 398B
- **Active per token:** ~13B
- **Base model:** Arcee Trinity-Large-Thinking
- **Architecture:** afmoe (SWA + global)
- **Experts:** 256 experts, 4 active + 1 shared
- **Target VRAM / RAM:** 48 GB / 64 GB
- **Tool calling:** not yet supported
- **Tier:** Paid
- **License:** Apache 2.0

## Source weights

The public GGUF this model will be packed from, pinned to a commit.

- **Repository:** [bartowski/arcee-ai_Trinity-Large-Thinking-GGUF](https://huggingface.co/bartowski/arcee-ai_Trinity-Large-Thinking-GGUF/tree/3854f407a7e4f2c025f8d55b9ed34fdc5970d085)
- **Revision:** `3854f407a7e4f2c025f8d55b9ed34fdc5970d085`
- **Quantization:** Q4_K_M
- **Source size:** 241.85 GB (7 files)

## Not yet verified

This model hasn't been benchmarked or verified on our hardware yet. Performance and hardware requirements are targets, not measured results; download sizes are measured wherever the model is already packed. No download or purchase until it clears the same verification the available models passed.

## Get the model

- **Download (target):** ~241.85 GB (single .lmpack, from Q4_K_M)