Instructions to use tomofusa/exp040-soup-3model-weighted with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tomofusa/exp040-soup-3model-weighted with Transformers:

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

pipe = pipeline("text-generation", model="tomofusa/exp040-soup-3model-weighted")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tomofusa/exp040-soup-3model-weighted")
model = AutoModelForCausalLM.from_pretrained("tomofusa/exp040-soup-3model-weighted")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tomofusa/exp040-soup-3model-weighted with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tomofusa/exp040-soup-3model-weighted"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tomofusa/exp040-soup-3model-weighted",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tomofusa/exp040-soup-3model-weighted

SGLang

How to use tomofusa/exp040-soup-3model-weighted 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 "tomofusa/exp040-soup-3model-weighted" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tomofusa/exp040-soup-3model-weighted",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "tomofusa/exp040-soup-3model-weighted" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tomofusa/exp040-soup-3model-weighted",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tomofusa/exp040-soup-3model-weighted with Docker Model Runner:
```
docker model run hf.co/tomofusa/exp040-soup-3model-weighted
```

exp040-soup-3model-weighted

Weighted Model Soup of 3 fine-tuned models for structured output generation (JSON / YAML / TOML / XML / CSV).

Full 16-bit merged weights. No adapter loading required.

Model Soup Configuration

This model is created by weighted averaging of 3 independently trained models:

Weight	Model	Training	Score
0.50	tomofusa/exp017-dpo-ipo-merged	SFT + DPO (IPO, lr=5e-7)	0.789
0.25	tomofusa/exp020-simpo-merged	SFT + CPO/SimPO (beta=2.5)	0.789
0.25	tomofusa/exp034-toml-upsample-dpo-merged	SFT (TOML upsampled) + DPO (IPO)	0.765

Soup method: model_A * 0.5 + model_B * 0.25 + model_C * 0.25 applied to all weight tensors.

Training Pipeline (per source model)

All source models share the same base pipeline:

Base model: Qwen/Qwen3-4B-Instruct-2507
SFT: QLoRA on structured output data (7,500 samples)
- SFT adapter: tomofusa/exp015-blend-h-lora
- Sources: daichira/structured-5k-mix-sft (5,000) + daichira/structured-hard-sft-4k (2,000 sampled) + custom TOML data (500)
- lr=5e-6, epochs=2, LoRA r=64/alpha=128, max_seq_len=1024
DPO: IPO/SimPO on u-10bei/dpo-dataset-qwen-cot (4,040 samples)
- lr=5e-7, beta=0.1, epochs=1, LoRA r=64/alpha=128
Merge + Soup: Each model merged to 16-bit, then weighted-averaged

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "tomofusa/exp040-soup-3model-weighted"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto",
)

Sources & Terms

Base model: Qwen/Qwen3-4B-Instruct-2507 - Apache 2.0
SFT data: daichira/structured-5k-mix-sft (CC-BY-4.0), daichira/structured-hard-sft-4k (CC-BY-4.0)
DPO data: u-10bei/dpo-dataset-qwen-cot
Users must comply with all upstream licenses and terms of use.

Downloads last month: 1

Safetensors

Model size

4B params

Tensor type

F32

Model tree for tomofusa/exp040-soup-3model-weighted

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

(1708)

this model

tomofusa
/

exp040-soup-3model-weighted