Downloads · 30 days
3
2% of all-time downloads
sasa5555/qwen3-4b-structured-output-lora_sft-creandata_merged
qwen3-4b-structured-output-lora_sft-creandata_merged is a text generation model from sasa5555. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This model is a merged version of the SFT (Supervised Fine-Tuning) adapter sasa5555/qwen3-4b-structured-output-lorasft-creandata with the base model Qwen/Qwen3-4B-Instruct-2507.
Downloads · 30 days
3
2% of all-time downloads
All-time downloads
142
Public
Parameters
4B
8.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8 GB · 100%
From the Hugging Face model README
This model is a merged version of the SFT (Supervised Fine-Tuning) adapter sasa5555/qwen3-4b-structured-output-lora_sft-creandata with the base model Qwen/Qwen3-4B-Instruct-2507.
This merged model combines the base model's capabilities with fine-tuned structured output generation abilities.
Since this is a merged model, you can use it directly with transformers without loading adapters:
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_id = "sasa5555/qwen3-4b-structured-output-lora_sft-creandata_merged"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype=torch.float16,
device_map="auto",
trust_remote_code=True,
)
# Test inference
prompt = "Convert this CSV to YAML: name,age\nJohn,30"
messages = [{"role": "user", "content": prompt}]
inputs = tokenizer.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
return_tensors="pt"
).to("cuda")
outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.0)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
This model was created by merging a LoRA adapter (trained via SFT) with the base model. The adapter was trained on structured output tasks including:
If you use this model, please cite: