---
title: WeDLM-8B-Base
canonical_url: "https://www.modelscope.ai/models/tencent-community/WeDLM-8B-Base"
md_url: "https://www.modelscope.ai/models/tencent-community/WeDLM-8B-Base.md"
repository: tencent-community/WeDLM-8B-Base
last_updated: 2026-10-10
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - wedlm
architectures:
  - WeDLMForCausalLM
base_model:
  - Qwen/Qwen3-8B
base_model_relation: finetune
parameters: 8.2B
tensor_type:
  - BF16
library_name:
  - transformer
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 38
stars: 0
tags:
  - "language model"
  - parallel-decoding
---

# WeDLM-8B-Base

> WeDLM-8B-Base - An open-source model by tencent-community on ModelScope. WeDLM-8B is a diffusion language model that performs parallel decoding under standard causal attention, initialized from Qwen3-8B.

tencent-community/WeDLM-8B-Base is a 8.2B-parameter text-generation model on ModelScope. licensed under apache-2.0. derived from Qwen/Qwen3-8B.

- **Repository**: tencent-community/WeDLM-8B-Base
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 8.2B
- **Base model**: Qwen/Qwen3-8B
- **Tags**: language model, parallel-decoding
- **Downloads**: 38
- **Stars**: 0
- **Last updated**: 2026-10-10

Source: https://www.modelscope.ai/models/tencent-community/WeDLM-8B-Base

---

# WeDLM-8B

**WeDLM-8B** is a diffusion language model that performs parallel decoding under standard causal attention, initialized from [Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B).

This is the **base (pretrained)** version. For the instruction-tuned version, see [WeDLM-8B-Instruct](https://huggingface.co/tencent/WeDLM-8B-Instruct).

📄 Paper (Coming Soon) | 🌐 [Project Page](https://wedlm.github.io) | 💻 [GitHub](https://github.com/tencent/WeDLM)

## Model Details

| Attribute | Value |
|:----------|:------|
| Initialized From | [Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) |
| Parameters | 8B |
| Context Length | 32,768 |

## Quick Start (Recommended)

For **fast inference**, use the `wedlm` engine:

```bash
pip install git+https://github.com/tencent/WeDLM.git
```

```python
from wedlm import LLM, SamplingParams

llm = LLM(model="tencent/WeDLM-8B")

prompt = "The theory of relativity states that"
outputs = llm.generate([prompt], SamplingParams(max_tokens=256))

print(outputs[0]["text"])
```

## HuggingFace Transformers

For **training** or simple forward passes, you can load via Transformers:

```python
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tencent/WeDLM-8B", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    "tencent/WeDLM-8B", 
    trust_remote_code=True,
    torch_dtype="auto",
    device_map="auto"
)

inputs = tokenizer("The theory of relativity", return_tensors="pt").to(model.device)
outputs = model(**inputs)
```

> ⚠️ **Note:** The HuggingFace interface is for training/forward pass convenience. For optimized inference throughput, use the `wedlm` engine above.

## Performance

| Benchmark | Qwen3-8B | WeDLM-8B |
|:----------|:--------:|:--------:|
| ARC-C (0-shot) | 92.66 | **92.92** |
| GSM8K (3-shot) | 85.97 | **90.20** |
| MATH (4-shot) | 50.80 | **53.60** |
| HumanEval (4-shot) | 68.90 | **75.00** |
| MMLU (5-shot) | 74.03 | **75.46** |
| **Average** | 72.61 | **74.72** |

## Citation (Coming soon)


## License

Apache 2.0