Instructions to use HEART77/TIR-Qwen3.5-4B-NonThinking-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HEART77/TIR-Qwen3.5-4B-NonThinking-v2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="HEART77/TIR-Qwen3.5-4B-NonThinking-v2")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("HEART77/TIR-Qwen3.5-4B-NonThinking-v2")
model = AutoModelForMultimodalLM.from_pretrained("HEART77/TIR-Qwen3.5-4B-NonThinking-v2", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use HEART77/TIR-Qwen3.5-4B-NonThinking-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HEART77/TIR-Qwen3.5-4B-NonThinking-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HEART77/TIR-Qwen3.5-4B-NonThinking-v2",
		"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/HEART77/TIR-Qwen3.5-4B-NonThinking-v2

SGLang

How to use HEART77/TIR-Qwen3.5-4B-NonThinking-v2 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 "HEART77/TIR-Qwen3.5-4B-NonThinking-v2" \
    --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": "HEART77/TIR-Qwen3.5-4B-NonThinking-v2",
		"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 "HEART77/TIR-Qwen3.5-4B-NonThinking-v2" \
        --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": "HEART77/TIR-Qwen3.5-4B-NonThinking-v2",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use HEART77/TIR-Qwen3.5-4B-NonThinking-v2 with Docker Model Runner:
```
docker model run hf.co/HEART77/TIR-Qwen3.5-4B-NonThinking-v2
```

TIR-Qwen3.5-4B-NonThinking-v2

Full-parameter SFT of Qwen/Qwen3.5-4B on the cleaned TIR mixture (HEART77/TIR-Training-Data, config v2_nonthink_clean, 25,065 rows). One contiguous epoch at a strict 32,768-token maximum length, trained 2026-09-02/03 on 4xA100 80GB with ms-swift 4.5.2 and Megatron Core 0.16.1.

This is the successor of HEART77/TIR-Qwen3.5-4B-NonThinking (trained on the legacy mixture). Compared with that run it drops the 2,200-row Cotrain attachment (10x image reuse, direct answers without reasoning), 20 rows whose final turn lacked <answer> tags, and four rows referencing intermediate images above PIL's decompression-bomb threshold. All other hyper-parameters are identical, so the two checkpoints form a data-only A/B pair. See the dataset card for the full audit.

Training data

Component	Rows
AgenticData true-multiturn SFT	4,537
CodeDance SFT (multi-turn `<code>` tool use)	13,828
TIR synthetic non-thinking attachment	6,700
Total	25,065

No assistant message contains a <think> block; the empty non-thinking prefix is inserted by the template and excluded from the loss.

Training configuration

Base model: Qwen/Qwen3.5-4B; full-parameter SFT, vision tower and aligner frozen
Recipe: non-thinking (add_non_thinking_prefix=true, loss_scale=default+ignore_empty_think, truncation_strategy=delete)
Learning rate 1e-5, min 1e-6, cosine decay, warmup fraction 0.05, weight decay 0.1, clip grad 1.0
Maximum length 32,768; global batch 64; micro batch 1; TP=2, PP=1, sequence parallel; bf16; flash attention
One epoch = 391/391 optimizer steps, elapsed 6h 12m 6s, final training loss 0.1767263
Full arguments: training/args.json; per-step log: training/logging.jsonl

Evaluation

Not yet evaluated at publication time. Results on TIR-Bench and the huawei_benchmark suite (FVQA, SimpleVQA, MMBrowseComp, GAIA) will be added alongside the legacy checkpoint for comparison.

Inference

vllm serve HEART77/TIR-Qwen3.5-4B-NonThinking-v2 --port 8000 --max-model-len 32768 --reasoning-parser qwen3

Disable thinking at request time with chat_template_kwargs={"enable_thinking": false}. The model uses the TIR agent format: Python in <code> blocks, tool results as a user turn beginning with Code execution result:, final answer in <answer>...</answer>.

Downloads last month: 21

Safetensors

Model size

5B params

Tensor type

BF16

Model tree for HEART77/TIR-Qwen3.5-4B-NonThinking-v2

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(784)

this model