README.md

metadata

license: apache-2.0

Important Links

📖Github | 🤖ModelScope | 🌐XiYan-SQL | 🌕析言GBI | 💻ModelScope Space

Introduction

We are excited to update our new XiYanSQL-QwenCoder series model, demonstrating improvements over its predecessor in some key features.

• The new XiYanSQL-QwenCoder model applies the merits of GRPO training strategy without thinking process, maintaining high efficiency and accuracy in SQL generation.
• The new XiYanSQL-QwenCoder model keeps its great performance in various benchmarks, including BIRD, Spider and DW benchmarks which will be released in the future.
• The new XiYanSQL-QwenCoder model demonstrates better generalization than its predecessor, especially in different dialects and out-of-domain datasets.

Model Downloads

Model	Download Latest
XiYanSQL-QwenCoder-7B	🤗HuggingFace 🤖Modelscope
XiYanSQL-QwenCoder-32B	🤗HuggingFace 🤖Modelscope

Performance

The XiYanSQL-QwenCoder models, as multi-dialect SQL base models, demonstrating robust SQL generation capabilities. The following presents the evaluation results at the time of release. We conducted a comprehensive evaluation of the model's performance under two schema formats, M-Schema, and original DDL, using the BIRD and Spider as SQLite benchmarks in the Text-to-SQL domain, as well as DW benchmarks for PostgreSQL and MySQL dialects.

Model name	Size	BIRD Dev@M-Schema	BIRD Dev@DDL	Spider Test@M-Schema	Spider Test@DDL	DW PostgreSQL@M-Schema	DW MySQL@M-Schema
GPT-4o-0806	UNK	58.47%	54.82%	82.89%	78.45%	46.79%	57.77%
GPT-4.1-0414	UNK	59.39%	54.11%	84.45%	79.86%	54.29%	63.18%
Claude3.5-sonnet-1022	UNK	53.32%	50.46%	76.27%	73.04%	55.22%	52.84%
Claude3.7-sonnet	UNK	54.82%	49.22%	78.04%	74.66%	53.23%	54.61%
Gemini-1.5-Pro	UNK	61.34%	57.89%	85.11%	84.00%	52.78%	62.78%
Gemini-2.5-Pro	UNK	67.21%	63.43%	88.29%	86.27%	63.16%	65.37%
DeepSeek-V2.5-1210	236B	55.74%	55.61%	82.08%	80.57%	45.74%	52.18%
DeepSeek-V3	685B	59.58%	56.71%	81.52%	79.91%	52.56%	55.95%
DeepSeek-R1	685B	58.15%	55.61%	80.72%	78.85%	60.56%	xx%
DeepSeek-R1-Distill-Qwen-32B	32B	50.65%	48.31%	78.65%	77.33%	37.22%	44.72%
Deepseek-Coder-33B-Instruct	33B	47.52%	44.72%	72.39%	xx%	31.48%	36.17%
OmniSQL-32B	32B	60.37%	55.87%	85.16%	83.19%	38.19%	42.34%
XiYanSQL-QwenCoder-32B-2412	32B	67.07%	63.04%	88.39%	85.46%	45.07%	52.84%
XiYanSQL-QwenCoder-32B-2504	32B	67.14%	62.26%	89.20%	86.17%	53.52%	57.74%

Requirements

transformers >= 4.37.0 vllm >= 0.7.2

Quickstart with Transformers and vLLM

Here is a simple code snippet for quickly using XiYanSQL-QwenCoder model. We provide a Chinese version of the prompt, and you just need to replace the placeholders for "question," "db_schema," and "evidence" to get started. We recommend using our M-Schema format for the schema; other formats such as DDL are also acceptable, but they may affect performance. Currently, we mainly support mainstream dialects like SQLite, PostgreSQL, and MySQL.

Prompt Template

nl2sqlite_template_cn = """你是一名{dialect}专家，现在需要阅读并理解下面的【数据库schema】描述，以及可能用到的【参考信息】，并运用{dialect}知识生成sql语句回答【用户问题】。
【用户问题】
{question}

【数据库schema】
{db_schema}

【参考信息】
{evidence}

【用户问题】
{question}

```sql"""

Inference with Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "XGenerationLab/XiYanSQL-QwenCoder-32B-2502"
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

tokenizer = AutoTokenizer.from_pretrained(model_name)

## dialects -> ['SQLite', 'PostgreSQL', 'MySQL']
prompt = nl2sqlite_template_cn.format(dialect="", db_schema="", question="", evidence="")
message = [{'role': 'user', 'content': prompt}]

text = tokenizer.apply_chat_template(
    message,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    pad_token_id=tokenizer.pad_token_id,
    eos_token_id=tokenizer.eos_token_id,
    max_new_tokens=1024,
    temperature=0.1,
    top_p=0.8,
    do_sample=True,
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Inference with vLLM

from vllm import LLM, SamplingParams
from transformers import AutoTokenizer
model_path = "XGenerationLab/XiYanSQL-QwenCoder-32B-2502"
llm = LLM(model=model_path, tensor_parallel_size=8)
tokenizer = AutoTokenizer.from_pretrained(model_path)
sampling_params = SamplingParams(
    n=1,
    temperature=0.1,
    max_tokens=2048
)

## dialects -> ['SQLite', 'PostgreSQL', 'MySQL']
prompt = nl2sqlite_template_cn.format(dialect="", db_schema="", question="", evidence="")
message = [{'role': 'user', 'content': prompt}]
text = tokenizer.apply_chat_template(
    message,
    tokenize=False,
    add_generation_prompt=True
)
outputs = llm.generate([text], sampling_params=sampling_params)
response = outputs[0].outputs[0].text

Acknowledgments

If you find our work useful, please give us a citation or a like, so we can make a greater contribution to the open-source community!