Instructions to use ModalityDance/I2S-Qwen3-8B-Confidence with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ModalityDance/I2S-Qwen3-8B-Confidence with Transformers:

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

pipe = pipeline("text-generation", model="ModalityDance/I2S-Qwen3-8B-Confidence")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ModalityDance/I2S-Qwen3-8B-Confidence")
model = AutoModelForCausalLM.from_pretrained("ModalityDance/I2S-Qwen3-8B-Confidence", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ModalityDance/I2S-Qwen3-8B-Confidence with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ModalityDance/I2S-Qwen3-8B-Confidence"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ModalityDance/I2S-Qwen3-8B-Confidence",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ModalityDance/I2S-Qwen3-8B-Confidence

SGLang

How to use ModalityDance/I2S-Qwen3-8B-Confidence 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 "ModalityDance/I2S-Qwen3-8B-Confidence" \
    --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": "ModalityDance/I2S-Qwen3-8B-Confidence",
		"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 "ModalityDance/I2S-Qwen3-8B-Confidence" \
        --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": "ModalityDance/I2S-Qwen3-8B-Confidence",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ModalityDance/I2S-Qwen3-8B-Confidence with Docker Model Runner:
```
docker model run hf.co/ModalityDance/I2S-Qwen3-8B-Confidence
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

I2S: Qwen3-8B Confidence

Overview

Intervention-to-Skill (I2S) turns large-model assistance into procedural skills that a small model can reuse across tasks. This repository provides the pretrained confidence model based on Qwen3-8B.

The model assesses a proposed environment action using the task, visible state, admissible actions, interaction history, and procedural guidance. Confidence is the normalized likelihood of CONFIDENT among the complete UNCERTAIN and CONFIDENT label sequences. This score guides when I2S requests large-model assistance.

I2S uses this model for confidence scoring while a separate actor generates actions. The model weights remain fixed during task streams.

Links

📄 Paper · 🤗 Hugging Face Paper · 🌐 Project Page · 💻 GitHub · 🧪 Usage Example · 📊 Evaluation · ModalityDance

Model Details

Setting	Value
Base model	Qwen/Qwen3-8B
Checkpoint format	Transformers, sharded Safetensors
Precision	Float32
Labels	`UNCERTAIN`, `CONFIDENT`

The repository contains the complete model weights, tokenizer, and configuration. Load the checkpoint directly with Transformers or vLLM.

Usage

Install the I2S package and model dependencies in a Python environment with CUDA:

git clone https://github.com/ModalityDance/I2S.git
cd I2S
python -m pip install -e '.[training]'

Load the pretrained model and score a proposed action:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

from i2s.confidence_scoring import encode_pair, label_log_likelihoods

model_id = "ModalityDance/I2S-Qwen3-8B-Confidence"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype=torch.float32, device_map="auto"
).eval()

messages = [{"role": "user", "content": (
    "Task: Move the item to the designated location.\n"
    "Observation: The item is held and the destination is accessible.\n"
    "Admissible actions: Place the item at the destination.\n"
    "Proposed action: Place the item at the destination."
)}]
prompt_ids, label_ids = encode_pair(tokenizer, messages)
with torch.inference_mode():
    scores = label_log_likelihoods(model, tokenizer, prompt_ids, label_ids)
    confidence = scores.softmax(0)[1].item()

print("Confidence:", confidence)

For task execution, follow the I2S Quick Start. Start the pretrained confidence model in the serving environment:

conda activate i2s-serving
python scripts/serve_models.py confidence

Training

The model is calibrated offline on interaction examples with labels derived from local execution feedback. Training and validation use disjoint task identities. The objective is class-weighted cross-entropy over the complete confidence-label likelihoods.

Training commands are in Confidence Calibration.

Citation

@misc{shi2026i2s,
  title  = {From Intervention to Skill: Evolving Small--Large Language Model Collaboration},
  author = {Zhenning Shi and Ran Wei and Ning Song and Yongqi Li},
  year   = {2026},
  url    = {https://github.com/ModalityDance/I2S}
}

License

This model is released under the Apache 2.0 License, following Qwen3-8B. The I2S code is released under the MIT License.

Downloads last month: 36

Safetensors

Model size

8B params

Tensor type

F32

Model tree for ModalityDance/I2S-Qwen3-8B-Confidence

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Finetuned

(2253)

this model

Quantizations

2 models