Rodimus+-Coder

🤖 ModelScope 🤗 Hugging Face 🖥️ GitHub

Introduction

Rodimus* is a new series of efficient large language models designed to address the challenges of computational complexity in Transformer-based architectures. The Rodimus* includes the base Rodimus model and its enhanced version, Rodimus+. Rodimus leverages a novel Data-Dependent Tempered Selection (DDTS) mechanism within a purely recurrent, linear attention-based framework, achieving high performance.

Building on this, Rodimus+ combines the strengths of Rodimus and the innovative Sliding Window Shared-Key Attention (SW-SKA) in a hybrid approach. This combination effectively integrates semantic, token, and head compression techniques, enabling a balance between accuracy and efficiency.

Beyond academic validation, we train and open-source the lightweight Rodimus+-Coder code LLM, based on the Rodimus architecture. It comes in sizes of 1.6B and 4B, and achieves outstanding results that surpass state-of-the-art (SOTA) models of the same size.

For more details, please refer to our Paper and Github.

Model Downloads

You can download the following table to see the various parameters for your use case. If you are located in mainland China, we also provide the model on modelscope.cn to speed up the download process.

Rodimus+-Coder Evaluation

We re-evaluate the metrics of the Qwen series models, and the metrics of other series models are quoted from the original paper. For detailed evaluation code, please refer to the evaluation method of Ling-Coder-Lite in CodeFuse-Evaluation.

Rodimus+-Coder-Base

Rodimus+-Coder-Chat

Usage

Installation

1. The latest version of transformers is recommended (at least 4.42.0).
2. We evaluate our models with python=3.8 and torch==2.1.2.
3. If you use Rodimus, you need to install flash-linear-attention, causal_conv1d and triton>=2.2.0. If you use Rodimus+, you need to further install flash-attention.

Generation

generate APi

import os
import torch
from modeling_rodimus import RodimusForCausalLM
from tokenization_rodimus_fast import RodimusTokenizer

# load model
ckpt_dir = "model_path"
tokenizer = RodimusTokenizer.from_pretrained(ckpt_dir)
model = RodimusForCausalLM.from_pretrained(
    ckpt_dir,
    torch_dtype=torch.bfloat16,
    device_map="cuda"
).eval()

# inference
input_prompt = "Write a quick sort algorithm in python."

messages = [
    {"role": "HUMAN", "content": input_prompt}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
)

model_inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**model_inputs, max_new_tokens=2048)
response = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]

print(response)

Citation

If you find our work helpful, feel free to give us a cite.

@inproceedings{
    he2025rodimus,
    title={Rodimus*: Breaking the Accuracy-Efficiency Trade-Off with Efficient Attentions},
    author={Zhihao He and Hang Yu and Zi Gong and Shizhan Liu and Jianguo Li and Weiyao Lin},
    booktitle={The Thirteenth International Conference on Learning Representations},
    year={2025},
    url={https://openreview.net/forum?id=IIVYiJ1ggK}
}