Instructions to use modrill/qwen3-4b-think-s1-ep23-full-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use modrill/qwen3-4b-think-s1-ep23-full-sft with Transformers:

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

pipe = pipeline("text-generation", model="modrill/qwen3-4b-think-s1-ep23-full-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("modrill/qwen3-4b-think-s1-ep23-full-sft")
model = AutoModelForCausalLM.from_pretrained("modrill/qwen3-4b-think-s1-ep23-full-sft", device_map="auto")
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 Settings

vLLM

How to use modrill/qwen3-4b-think-s1-ep23-full-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "modrill/qwen3-4b-think-s1-ep23-full-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "modrill/qwen3-4b-think-s1-ep23-full-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/modrill/qwen3-4b-think-s1-ep23-full-sft

SGLang

How to use modrill/qwen3-4b-think-s1-ep23-full-sft 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 "modrill/qwen3-4b-think-s1-ep23-full-sft" \
    --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": "modrill/qwen3-4b-think-s1-ep23-full-sft",
		"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 "modrill/qwen3-4b-think-s1-ep23-full-sft" \
        --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": "modrill/qwen3-4b-think-s1-ep23-full-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use modrill/qwen3-4b-think-s1-ep23-full-sft with Docker Model Runner:
```
docker model run hf.co/modrill/qwen3-4b-think-s1-ep23-full-sft
```

Qwen3-4B Think S1 Ep23 (Full SFT)

Full-parameter supervised fine-tuning (SFT) of Qwen/Qwen3-4B-Base on the ocr_think_50k dataset with the Qwen3 chat template (think-style reasoning).

This checkpoint is stage 1 episode 23: training continued from an internal think_s1 run at checkpoint-302 (same base architecture), then fine-tuned for two additional epochs on ocr_think_50k.

Model description

Method: full SFT (all weights trainable), DeepSpeed ZeRO-3, 4 GPUs
Dataset: ocr_think_50k
Template: qwen3
Not LoRA / not QLoRA: entire 4B model was updated

Training details

Field	Value
Epochs	2
Seed	42
cutoff_len	24576
packing	true
neat_packing	false
per_device_train_batch_size	1
gradient_accumulation_steps	16
effective_batch_size	64
learning_rate	5e-5
train_loss	0.5416
train_steps	604
finished_at	2026-06-10 05:23 CST

Optimizer: AdamW (fused), cosine schedule, warmup ratio 0.1. Framework: Transformers 5.6.0, PyTorch 2.8.0+cu128.

Related models

No-think SFT (same project): modrill/qwen3-4b-nothink-s1-full-sft - OpenCodeInstruct, qwen3_nothink template
Base: Qwen/Qwen3-4B-Base

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

repo_id = "modrill/qwen3-4b-think-s1-ep23-full-sft"
tokenizer = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo_id,
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)

License

Released under Apache 2.0 (see LICENSE in the upstream Qwen model card if not bundled here).

Downloads last month: 9

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for modrill/qwen3-4b-think-s1-ep23-full-sft

Base model

Qwen/Qwen3-4B-Base

Finetuned

(423)

this model