Text Generation
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.

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}, 
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for leonardklin/B3D-RWKV

Base model

BlinkDL/rwkv7-g1
Finetuned
(19)
this model

Datasets used to train leonardklin/B3D-RWKV

Collection including leonardklin/B3D-RWKV

Paper for leonardklin/B3D-RWKV