Instructions to use kiratan/qwen3-4b-structeval-lora-90 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kiratan/qwen3-4b-structeval-lora-90 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit")
model = PeftModel.from_pretrained(base_model, "kiratan/qwen3-4b-structeval-lora-90")

Transformers

How to use kiratan/qwen3-4b-structeval-lora-90 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="kiratan/qwen3-4b-structeval-lora-90")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("kiratan/qwen3-4b-structeval-lora-90", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use kiratan/qwen3-4b-structeval-lora-90 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kiratan/qwen3-4b-structeval-lora-90"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kiratan/qwen3-4b-structeval-lora-90",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kiratan/qwen3-4b-structeval-lora-90

SGLang

How to use kiratan/qwen3-4b-structeval-lora-90 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "kiratan/qwen3-4b-structeval-lora-90" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kiratan/qwen3-4b-structeval-lora-90",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "kiratan/qwen3-4b-structeval-lora-90" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kiratan/qwen3-4b-structeval-lora-90",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Studio new

How to use kiratan/qwen3-4b-structeval-lora-90 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kiratan/qwen3-4b-structeval-lora-90 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kiratan/qwen3-4b-structeval-lora-90 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for kiratan/qwen3-4b-structeval-lora-90 to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="kiratan/qwen3-4b-structeval-lora-90",
    max_seq_length=2048,
)

Docker Model Runner
How to use kiratan/qwen3-4b-structeval-lora-90 with Docker Model Runner:
```
docker model run hf.co/kiratan/qwen3-4b-structeval-lora-90
```

qwen3-4b-structured-output-lora

This repository provides a LoRA adapter fine-tuned from Qwen/Qwen3-4B-Instruct-2507 using QLoRA (4-bit, Unsloth).

This repository contains LoRA adapter weights only. The base model must be loaded separately.

Training Objective

This adapter is trained to improve structured output accuracy (JSON / YAML / XML / TOML / CSV).

Loss is applied only to the final assistant output, while intermediate reasoning (Chain-of-Thought) is masked.

Training Configuration

Base model: Qwen/Qwen3-4B-Instruct-2507
Method: QLoRA (4-bit)
Max sequence length: 2048
Epochs: 1
Learning rate: 1e-06
LoRA: r=64, alpha=128

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = "your_id/your-repo"

tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
    base,
    torch_dtype=torch.float16,
    device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)

Sources & Terms (IMPORTANT)

Training data: kiratan/daichira-12k

Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License. Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.

Framework versions

PEFT 0.18.1

Downloads last month: -

kiratan
/

qwen3-4b-structeval-lora-90

Training Objective

Training Configuration

Usage

Sources & Terms (IMPORTANT)

Framework versions

Dataset used to train kiratan/qwen3-4b-structeval-lora-90