Downloads · 30 days
28
8% of all-time downloads
I-DLM-8B-lora-r128
yifanyu/I-DLM-8B-lora-r128
I-DLM-8B-lora-r128 is a text generation model from yifanyu. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
LoRA adapter (rank=128) for I-DLM-8B, enabling lossless Relaxed Introspective Strided Decoding (R-ISD).
All-time downloads
344
Public
Repo size
698 MB
Likes
2
Public
Repo makeup
Click a slice to open those files.
.safetensors698 MB · 99%
At a glance
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- Apr 7, 2026
- Updated
- Apr 15, 2026
- SHA
- ea5db26b
Try a prompt
Base models
Model card
From the Hugging Face model README
I-DLM-8B-LoRA-r128
LoRA adapter (rank=128) for I-DLM-8B, enabling lossless Relaxed Introspective Strided Decoding (R-ISD).
This model was presented in the paper Introspective Diffusion Language Models.
Overview
This LoRA adapter is used in the gated LoRA setup described in the I-DLM paper:
- LoRA is active only on MASK positions during ISD decoding
- Clean/verify positions use base model weights only
- This produces output bit-for-bit identical to the base AR model (Qwen3-8B)
Usage
Note: This LoRA adapter is used with the I-DLM-8B base model via our SGLang-based serving pipeline. Direct loading via
transformersis not currently supported for reproducing paper results. Please use the SGLang-based ISD pipeline for inference.
`
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Languages
- en
- Created
- Apr 7, 2026
- Updated
- Apr 15, 2026