---
license: apache-2.0
language:
- en
- es
- fr
- de
- it
- pt
- ru
- ar
- hi
- ko
- zh
library_name: transformers
base_model:
- arcee-ai/Trinity-Large-Preview
base_model_relation: quantized
---
# Trinity-Large-Preview-FP8-Block
## Introduction
Trinity-Large-Preview is a 398B-parameter sparse Mixture-of-Experts (MoE) model with approximately 13B active parameters per token. It is the largest model in Arcee AI's Trinity family, trained on more than 17 trillion tokens and delivering frontier-level performance with strong long-context comprehension.
Trinity-Large-Preview is a lightly post-trained model based on Trinity-Large-Base.
**This repository contains the FP8 block-quantized weights of Trinity-Large-Preview (FP8 weights and activations with per-block scaling).**
Try it at [chat.arcee.ai](http://chat.arcee.ai/)
More details on the training of Trinity Large are available in the [technical report](https://arxiv.org/abs/2602.17004).
## Quantization Details
- **Scheme:** `FP8 Block` (FP8 weights and activations, per-block scaling with E8M0 scale format)
- **Format:** `compressed-tensors`
- **Intended use:** High-throughput FP8 deployment of Trinity-Large-Preview with near-lossless quality, optimized for NVIDIA Hopper/Blackwell GPUs
- **Supported backends:** [DeepGEMM](https://github.com/deepseek-ai/DeepGEMM), vLLM CUTLASS, Triton
## Model Variants
The Trinity Large family consists of three checkpoints from the same training run:
- **[Trinity-Large-Preview](https://huggingface.co/arcee-ai/Trinity-Large-Preview)**: Lightly post-trained, chat-ready model undergoing active RL
- **[Trinity-Large-TrueBase](https://huggingface.co/arcee-ai/Trinity-Large-TrueBase)**: 10T-token pre-anneal pretraining checkpoint
- **[Trinity-Large-Base](https://huggingface.co/arcee-ai/Trinity-Large-Base)**: Full 17T-token pretrained foundation model with mid-training anneals
## Architecture
Trinity-Large-Preview uses a sparse MoE configuration designed to maximize efficiency while maintaining large-scale capacity.
| Hyperparameter | Value |
|:---|:---:|
| Total parameters | ~398B |
| Active parameters per token | ~13B |
| Experts | 256 (1 shared) |
| Active experts | 4 |
| Routing strategy | 4-of-256 (1.56% sparsity) |
| Dense layers | 6 |
| Pretraining context length | 8,192 |
| Context length after extension | 512k |
| Architecture | Sparse MoE (AfmoeForCausalLM) |
## Benchmarks
| Benchmark | Llama 4 Maverick | Trinity-Large Preview |
|-----------|------------------|----------------------|
| MMLU | 85.5 | 87.2 |
| MMLU-Pro | 80.5 | 75.2 |
| GPQA-Diamond | 69.8 | 63.3 |
| AIME 2025 | 19.3 | 24.0 |
## Training Configuration
### Pretraining
- Training tokens: 17 trillion
- Data partner: [Datology](https://www.datologyai.com/)
## Posttraining
- This checkpoint was instruction tuned on 20B tokens.
### Infrastructure
- Hardware: 2,048 NVIDIA B300 GPUs
- Parallelism: HSDP + Expert Parallelism
- Compute partner: [Prime Intellect](https://www.primeintellect.ai/)
## Usage
### Running our model
- [VLLM](https://huggingface.co/arcee-ai/Trinity-Large-Preview-FP8-Block#vllm)
- [Transformers](https://huggingface.co/arcee-ai/Trinity-Large-Preview-FP8-Block#transformers)
- [API](https://huggingface.co/arcee-ai/Trinity-Large-Preview-FP8-Block#api)
### Inference tested on
- 8x NVIDIA H100 80GB (tensor parallel = 8)
- vLLM 0.18.0+
### VLLM
Supported in VLLM release 0.18.0+ with DeepGEMM FP8 MoE acceleration.
```bash
# pip
pip install "vllm>=0.18.0"
```
Serving the model with DeepGEMM enabled:
```bash
VLLM_USE_DEEP_GEMM=1 vllm serve arcee-ai/Trinity-Large-Preview-FP8-Block \
--trust-remote-code \
--tensor-parallel-size 8 \
--enable-auto-tool-choice \
--tool-call-parser hermes
```
Serving without DeepGEMM (falls back to CUTLASS/Triton):
```bash
vllm serve arcee-ai/Trinity-Large-Preview-FP8-Block \
--trust-remote-code \
--tensor-parallel-size 8 \
--enable-auto-tool-choice \
--tool-call-parser hermes
```
### Transformers
Use the `main` transformers branch or pass `trust_remote_code=True` with a released version.
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
model_id = "arcee-ai/Trinity-Large-Preview-FP8-Block"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
device_map="auto",
trust_remote_code=True
)
messages = [
{"role": "user", "content": "Who are you?"},
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
outputs = model.generate(
input_ids,
max_new_tokens=256,
do_sample=True,
temperature=0.8,
top_k=50,
top_p=0.8
)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)
```
### API
Available on OpenRouter:
```bash
curl -X POST "https://openrouter.ai/v1/chat/completions" \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "arcee-ai/trinity-large-preview",
"messages": [
{
"role": "user",
"content": "What are some fun things to do in New York?"
}
]
}'
```
## License
Trinity-Large-Preview-FP8-Block is released under the Apache License, Version 2.0.
## Citation
If you use this model, please cite:
```bibtex
@misc{singh2026arceetrinity,
title = {Arcee Trinity Large Technical Report},
author = {Varun Singh and Lucas Krauss and Sami Jaghouar and Matej Sirovatka and Charles Goddard and Fares Obied and Jack Min Ong and Jannik Straube and Fern and Aria Harley and Conner Stewart and Colin Kealty and Maziyar Panahi and Simon Kirsten and Anushka Deshpande and Anneketh Vij and Arthur Bresnu and Pranav Veldurthi and Raghav Ravishankar and Hardik Bishnoi and DatologyAI Team and Arcee AI Team and Prime Intellect Team and Mark McQuade and Johannes Hagemann and Lucas Atkins},
year = {2026},
eprint = {2602.17004},
archivePrefix= {arXiv},
primaryClass = {cs.LG},
doi = {10.48550/arXiv.2602.17004},
url = {https://arxiv.org/abs/2602.17004}
}
```
