Instructions to use TIDE-dllm/distill-LLaDA2-TIDE_Shared with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TIDE-dllm/distill-LLaDA2-TIDE_Shared with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="TIDE-dllm/distill-LLaDA2-TIDE_Shared", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForMaskedLM
model = AutoModelForMaskedLM.from_pretrained("TIDE-dllm/distill-LLaDA2-TIDE_Shared", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TIDE-dllm/distill-LLaDA2-TIDE_Shared with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TIDE-dllm/distill-LLaDA2-TIDE_Shared"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TIDE-dllm/distill-LLaDA2-TIDE_Shared",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TIDE-dllm/distill-LLaDA2-TIDE_Shared

SGLang

How to use TIDE-dllm/distill-LLaDA2-TIDE_Shared with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "TIDE-dllm/distill-LLaDA2-TIDE_Shared" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TIDE-dllm/distill-LLaDA2-TIDE_Shared",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "TIDE-dllm/distill-LLaDA2-TIDE_Shared" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TIDE-dllm/distill-LLaDA2-TIDE_Shared",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TIDE-dllm/distill-LLaDA2-TIDE_Shared with Docker Model Runner:
```
docker model run hf.co/TIDE-dllm/distill-LLaDA2-TIDE_Shared
```

distill-LLaDA2-TIDE_Shared

This model was introduced in the paper Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models.

distill-LLaDA2-TIDE_Shared is a 0.6B diffusion language model distilled from LLaDA2.0-mini (16B MoE) into the Qwen3-0.6B-diffusion-bd3lm-v0.1 student in the Cross-Tokenizer (Pipeline A) of the TIDE framework. TIDAL + CompDemo applied within the cross-tokenizer pipeline (non-native ablation).

Model Overview

Method: TIDE — Reverse CALM / TIDAL / CompDemo (cross-architecture distillation for diffusion LMs)
Framework: TIDE / dLLM
Student (initialization): Qwen3-0.6B-diffusion-bd3lm-v0.1 (BD3LM, block_size=32)
Teacher: inclusionAI/LLaDA2.0-mini
Distillation mode: --distill_mode alm_taid --use_comp_demo True
Datasets: tulu-3-sft-mixture, smoltalk, opc-sft-stage1 and opc-sft-stage2 — same composition as the Qwen3-0.6B-diffusion-bd3lm-v0.1 base. Pre-tokenized for this teacher in TIDE-dllm/distill_llada2_sft.

Installation

pip install torch transformers accelerate

Quick Start

This checkpoint is fully compatible with the BD3LM generate(...) routine published with dllm-hub/Qwen3-0.6B-diffusion-bd3lm-v0.1 — only the model name changes.

import torch
from transformers import AutoModelForMaskedLM, AutoTokenizer

repo = "TIDE-dllm/distill-LLaDA2-TIDE_Shared"
device = "cuda" if torch.cuda.is_available() else "cpu"

model = AutoModelForMaskedLM.from_pretrained(
    repo, dtype=torch.bfloat16, trust_remote_code=True,
).to(device).eval()
tokenizer = AutoTokenizer.from_pretrained(repo, trust_remote_code=True)

prompts = [
    [
        {"role": "system", "content": "You are a helpful AI assistant."},
        {"role": "user", "content": "Implement a DFS traversal in Python with clear inline comments."},
    ],
]
encoded = [tokenizer.apply_chat_template(m, add_generation_prompt=True, tokenize=True, enable_thinking=False) for m in prompts]
# ... use the same `generate()` function as in dllm-hub/Qwen3-0.6B-diffusion-bd3lm-v0.1.

Command-Line Interface

For an interactive demo (visualised iterative denoising), use the script in the TIDE / dLLM repo:

python -u examples/a2d/bd3lm/chat.py \
    --model_name_or_path TIDE-dllm/distill-LLaDA2-TIDE_Shared \
    --chat_template True --block_size 32 --remasking low_confidence \
    --steps 256 --max_new_tokens 256

Reproducing this checkpoint

git clone https://github.com/PKU-YuanGroup/TIDE && cd TIDE
pip install -e . && git submodule update --init --recursive
pip install -e "lm-evaluation-harness[ifeval,math]" && pip install -e "tokenkit[full]"

# Download the pre-tokenized SFT mixture for this teacher
huggingface-cli download TIDE-dllm/distill_llada2_sft --repo-type dataset \
    --local-dir data/distill_llada2_sft

bash scripts/distill_llada2.sh \
    --data_path data/distill_llada2_sft \
    --distill_mode alm_taid --use_comp_demo True \
    --num_gpus 8

Citation

@misc{zhang2026turningtidecrossarchitecturedistillation,
      title={Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models},
      author={Gongbo Zhang and Wen Wang and Ye Tian and Li Yuan},
      year={2026},
      eprint={2604.26951},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2604.26951},
}