Instructions to use Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled with Transformers:

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

pipe = pipeline("image-text-to-text", model="Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled")
model = AutoModelForMultimodalLM.from_pretrained("Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled

SGLang

How to use Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled 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 "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled" \
    --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": "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled" \
        --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": "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled with Docker Model Runner:
```
docker model run hf.co/Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Qwen3.5-0.8B-Strata-distilled

Qwen/Qwen3.5-0.8B fine-tuned on Strata-distilled, a small synthetic chat dataset with reasoning traces at several thinking levels (off / low / medium / high).

Test release. This is an experiment to test a distillation pipeline, not a finished model.

Disclaimer

The training data was generated by swift-1.5-iq3_xxs, a heavily quantized model, and was not checked before training. Expect factual errors, flawed reasoning and uneven quality.
Do not rely on this model for anything important.

Training


Method	LoRA (r=64), merged into the weights
Training examples	65 (one per assistant turn)
Epochs / steps	2 / 22
Learning rate	0.0002
Max sequence length	4096 tokens
Final training loss	1.4175
Precision	16-bit (bf16/fp16), base model not quantized

Only the assistant's replies (including <think> reasoning) were trained on.

Usage

Thinking is controlled by the chat template, as in the base model:

from transformers import AutoTokenizer, AutoModelForCausalLM

repo = "Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, device_map="auto")

messages = [{"role": "user", "content": "what's 17 times 23?"}]
prompt = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=True)  # False = no reasoning
inputs = tok(prompt, return_tensors="pt", add_special_tokens=False).to(model.device)
out = model.generate(**inputs, max_new_tokens=512)
print(tok.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Small models can loop in thinking mode, so set a max_new_tokens limit.

Downloads last month: 37

Safetensors

Model size

0.9B params

Tensor type

F32

BF16

Model tree for Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled

Base model

Qwen/Qwen3.5-0.8B-Base

Finetuned

Qwen/Qwen3.5-0.8B

Finetuned

(606)

this model

Quantizations

1 model

Stormtrooperaim
/

Qwen3.5-0.8B-Strata-distilled

Qwen3.5-0.8B-Strata-distilled

Disclaimer

Training

Usage

Model tree for Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled

Dataset used to train Stormtrooperaim/Qwen3.5-0.8B-Strata-distilled