metadata
base_model: BlinkDL/rwkv7-g1
datasets:
- allenai/tulu-3-sft-mixture
- Jackrong/GLM-5.1-Reasoning-1M-Cleaned
- angrygiraffe/claude-opus-4.6-4.7-reasoning-8.7k
language:
- en
- zh
- fr
- es
- de
- pt
- ru
- it
- ja
- ko
- vi
- ar
license: apache-2.0
pipeline_tag: text-generation
tags:
- dllm
- diffusion
- rwkv
- llm
Triplet-Block Diffusion RWKV
This repository contains the checkpoint of B3D-RWKV, a 7.2B-parameter RWKV language model presented in the paper Triplet-Block Diffusion RWKV.
B3D-RWKV is a diffusion RWKV variant that integrates the model's $O(L)$ inference efficiency with parallel, bidirectional discrete-diffusion through a triplet-block layout method. It reaches comparable accuracy on an 8-task suite versus existing models while significantly outperforming baselines in decoding throughput with an average of 1.6× speedup.
- Paper: Triplet-Block Diffusion RWKV
- Code: GitHub Repository
Usage
For usage, please see the B3D-RWKV infer and serve directories in the official repository for instructions on how to run inference and serve the model.
Note: This checkpoint is a supervised fine-tuned (SFT) version of rwkv7-g1f-7.2B.
Citation
@misc{lin2026tripletblockdiffusionrwkv,
title={Triplet-Block Diffusion RWKV},
author={Ke Lin and Yiyang Luo and Zhaolong Su and Yunya Song and Anyi Rao},
year={2026},
eprint={2605.25969},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2605.25969},
}