yuyuzhang's picture
Update README.md
f77b9ab verified
|
raw
history blame
5.74 kB
metadata
license: mit
base_model:
  - ByteDance-Seed/Seed-Coder-8B-Base

Seed-Coder-8B-Reasoning

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.

This repo contains Seed-Coder-8B-Reasoning model, which has the following features:

  • Type: Causal language models
  • Training Stage: Pretraining & Post-training
  • Data Source: Public datasets
  • Context Length: 32,768

Model Downloads

Model Name Length Download Notes
Seed-Coder-8B-Base 32K 🤗 Model Pretrained on our model-centric code data.
Seed-Coder-8B-Instruct 32K 🤗 Model Instruction-tuned for alignment with user intent.
👉 Seed-Coder-8B-Reasoning 32K 🤗 Model RL trained to boost reasoning capabilities.

Requirements

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

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:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

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

tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)

messages = [
    {"role": "user", "content": "Write a quick sort algorithm."},
]

input_ids = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    return_tensors="pt",
    add_generation_prompt=True,  
).to(model.device)

outputs = model.generate(input_ids, max_new_tokens=16384)
response = tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True)
print(response)

Evaluation

Seed-Coder-8B-Reasoning has been evaluated extensively on reasoning-intensive code benchmarks, showing:

  • Significant improvements on competitive programming datasets and coding challenges.
  • Enhanced ability to break down complex problems, design correct algorithms, and produce efficient implementations.
  • Strong generalization to unseen problems across multiple domains (math, strings, arrays, graphs, DP, etc.).
Model LiveCodeBench-Hard LiveCodeBench-Medium LiveCodeBench-Easy Overall
4mon3mon2mon 4mon3mon2mon 4mon3mon2mon
~8B Models
DeepSeek-R1-Distill-Qwen-7B 11.310.79.6 39.637.237.1 76.277.167.1 36.5
DeepSeek-R1-Distill-Seed-Coder-8B 13.613.913.4 39.638.739.3 79.880.273.2 39.0
OlympicCoder-7B 12.711.812.5 40.839.038.7 78.077.167.8 37.9
Qwen3-8B-thinking 27.523.519.7 65.759.758.5 98.098.197.3 57.4
Seed-Coder-8B-Reasoning 27.628.031.0 65.859.257.5 87.888.080.1 53.6
13B+ Models
DeepSeek-R1-Distill-Qwen-14B 21.320.516.1 58.153.451.4 93.394.293.7 51.9
Claude-3.7-Sonnet-thinking 27.330.831.0 54.555.151.4 96.2100.0100.0 53.3
o3-mini-low 30.332.328.6 69.661.254.1 98.7100.0100.0 59.4

For detailed benchmark performance, please refer to our 📑 Technical Report.