README.md

---
license: mit
---

# Seed-Coder-8B-Base

## Introduction
We are thrilled to introduce Seed-Coder, a powerful, transparent, and parameter-efficient family of open-source code models at the 8B scale, featuring base, instruct, and reasoning variants. Seed-Coder contributes to promote the evolution of open code models through the following highlights.

- Model-centric: Seed-Coder predominantly leverages LLMs instead of hand-crafted rules for code data filtering, minimizing manual effort in pretraining data construction.
- Transparent: We openly share detailed insights into our model-centric data pipeline, including methods for curating GitHub data, commits data, and code-related web data.
- Powerful: Seed-Coder achieves state-of-the-art performance among open-source models of comparable size across a diverse range of coding tasks.

<p align="center">
  <img width="100%" src="imgs/seed-coder_intro_performance.jpg">
</p>

## Highlight


Seed-Coder-8B-Base is an 8-billion-parameter foundation model tailored for code understanding and generation. It is designed to provide developers with a powerful, general-purpose code model capable of handling a wide range of coding tasks. It features: 
- Pretrained on a massively curated corpus, filtered using **LLM-based techniques** to ensure high-quality real-world code, resulting in cleaner and more effective learning signals.
- Excels at code completion and supports Fill-in-the-Middle (FIM) tasks, enabling it to predict missing code spans given partial contexts.
- Robust performance across various programming languages, making it ideal for downstream finetuning or direct use in code generation systems.
- Long-context support up to 32K tokens, enabling it to handle large codebases, multi-file projects, and extended editing tasks.

Seed-Coder-8B-Base serves as the foundation for Seed-Coder-8B-Instruct and Seed-Coder-8B-reasoning.



## Model Downloads
| Model Name                  | Length | Download   |    Notes |
|---------------------------------------------------------|--------|------------------------------------|-----------------------|
| 👉 **Seed-Coder-8B-Base**           | 32K    | 🤗 [Model](https://huggingface.co/ByteDance-Seed/Seed-Coder-8B-Base)   |  Pretrained on our model-centric code data.  |
| Seed-Coder-8B-Instruct             | 32K    | 🤗 [Model](https://huggingface.co/ByteDance-Seed/Seed-Coder-8B-Instruct)   |  Instruction-tuned for alignment with user intent. |
| Seed-Coder-8B-Reasoning            | 32K    | 🤗 [Model](https://huggingface.co/ByteDance-Seed/Seed-Coder-8B-Reasoning)   |  RL trained to boost reasoning capabilities.  |


## Requirements
You will need to install the latest versions of `transformers` and `accelerate`:

```bash
pip install -U transformers accelerate
```

## Quickstart

Here is a simple example demonstrating how to load the model and perform code generation using the Hugging Face `pipeline` API:

```python
import transformers
import torch

model_id = "ByteDance-Seed/Seed-Coder-8B-Base"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

output = pipeline("def say_hello_world():", max_new_tokens=100)
print(output[0]["generated_text"])
```

### Fill-in-the-Middle (FIM) Example

Seed-Coder-8B-Base natively supports **Fill-in-the-Middle (FIM)** tasks, where the model is given a prefix and a suffix and asked to predict the missing middle content.  
This allows for code infilling scenarios such as completing a function body or inserting missing logic between two pieces of code.

A typical usage flow:

```python
import transformers
import torch

model_id = "ByteDance-Seed/Seed-Coder-8B-Base"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

# You can concatenate a prefix, a special FIM separator token, and a suffix
prefix = "def add_numbers(a, b):\n    "
suffix = "\n    return result"

# Combine prefix and suffix following the FIM format
fim_input = '<[fim-suffix]>' + suffix + '<[fim-prefix]>' + prefix + '<[fim-middle]>'

output = pipeline(fim_input, max_new_tokens=512)
print(output[0]["generated_text"])
```

## Evaluation

Seed-Coder-8B-Base has been evaluated across a variety of code understanding and generation benchmarks.  
It demonstrates strong capabilities in:
- Fluent and contextually appropriate code completion.
- Reasoning about code structure and inferring missing logic.
- Generalizing across different programming languages, coding styles, and codebases.

|            | DeepSeek-Coder-6.7B-Base | OpenCoder-8B-Base | Qwen2.5-Coder-7B | Seed-Coder-8B-Base |
|------------|:------------------------:|:-----------------:|:----------------:|:------------------:|
| HumanEval  |           47.6           |        66.5       |       72.0       |        77.4        |
| MBPP       |           70.2           |        79.9       |       79.4       |        82.0        |
| MultiPL-E  |           44.7           |        61.0       |       58.8       |        67.6        |
| CruxEval-O |           41.0           |        43.9       |       56.0       |        48.4        |

For detailed benchmark performance, please refer to our [📑 technical report](https://github.com/ByteDance-Seed/Seed-Coder/blob/master/Seed-Coder.pdf).

<!-- ## Citation

If you find Seed-Coder helpful, please consider citing our work:

```
@article{zhang2025seedcoder,
    title={Seed-Coder: Let the Code Model Curate Data for Itself},
    author={Xxx},
    year={2025},
    eprint={2504.xxxxx},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/xxxx.xxxxx}, 
}
``` -->