GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Rubin-Wei
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
Rubin-Wei
Model Tree
Input Modalities
Output Modalities
Supported Functionality
This repository contains the 1.7B biology Memory Decoder released with Memory Decoder at Scale. It is a pretrained parametric long-term memory that can be swapped into a compatible frozen language-model backbone.
This 1.7B memory was trained on the biology CPT corpus using retrieval-derived sparse target distributions. The released dataset contains the aligned CPT text, Qwen3-preprocessed data, and KNN distributions. In the paper, the 1.7B Qwen3-vocabulary domain memories are evaluated with frozen Qwen3 Base backbones from 0.6B to 14B.
This checkpoint is a memory component, not a standalone chat- or instruction-tuned model. The memory and backbone must use compatible token IDs and vocabularies.
| Field | Value |
|---|---|
| Memory size | 1.7B class |
| Architecture/tokenizer family | Qwen3ForCausalLM memory; Qwen3 tokenizer/vocabulary |
| Domain | Biology |
| Evaluation benchmark | BioInst |
| Intended backbone | Frozen, tokenizer-compatible Qwen3 Base model |
| Release contents | Inference weights, configuration, and tokenizer files |
Install the matching environment from
LUMIA-Group/MemoryDecoder-at-Scale, then
set MODEL_PATH to a compatible frozen backbone and MEMDEC_PATH to this
repository:
bash
MODEL_PATH=/path/to/compatible-base-model \MEMDEC_PATH=Rubin-Wei/MemoryDecoder-Qwen3-1.7B-biology \bash eval/opencompass/scripts/domain/evaluate_bioinst.sh
See the repository documentation and launcher for benchmark-specific options, including interpolation weights and batch settings.
This checkpoint is intended for research and evaluation in the biology domain. Its outputs depend on the backbone, prompt, and interpolation settings. Domain specialization does not guarantee factual correctness or safety, and the model may inherit biases and errors from its training sources.
If you use this checkpoint, please cite:
bibtex
@misc{wei2026memorydecoderscalepretrained,title={Memory Decoder at Scale: A Pretrained, Parametric Long-Term Memory},author={Rubin Wei and Jiaqi Cao and Jiarui Wang and Junming Zhang and Qipeng Guo and Bowen Zhou and Zhouhan Lin},year={2026},eprint={2607.27919},archivePrefix={arXiv},primaryClass={cs.CL},url={https://arxiv.org/abs/2607.27919},}