Instructions to use mobilint/Qwen2.5-3B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mobilint/Qwen2.5-3B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="mobilint/Qwen2.5-3B-Instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("mobilint/Qwen2.5-3B-Instruct", trust_remote_code=True, device_map="auto")

Mobilint

How to use mobilint/Qwen2.5-3B-Instruct with Mobilint:

# pip install mblt-model-zoo
from mblt_model_zoo.vision import MBLT_Engine

model = MBLT_Engine(
    model_cls="Qwen2.5-3B-Instruct",
    model_type="DEFAULT",
    model_path="",
    core_mode="global8",
)

try:
    image = model.preprocess("path/to/image.jpg")
    output = model(image)
    result = model.postprocess(output)
finally:
    model.dispose()

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mobilint/Qwen2.5-3B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mobilint/Qwen2.5-3B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mobilint/Qwen2.5-3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mobilint/Qwen2.5-3B-Instruct

SGLang

How to use mobilint/Qwen2.5-3B-Instruct 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 "mobilint/Qwen2.5-3B-Instruct" \
    --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": "mobilint/Qwen2.5-3B-Instruct",
		"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 "mobilint/Qwen2.5-3B-Instruct" \
        --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": "mobilint/Qwen2.5-3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mobilint/Qwen2.5-3B-Instruct with Docker Model Runner:
```
docker model run hf.co/mobilint/Qwen2.5-3B-Instruct
```

Qwen2.5-3B-Instruct

File size: 1,018 Bytes

ca76d7b

{
  "architectures": [
    "MobilintQwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "proxy_qwen2.MobilintQwen2Config",
    "AutoModelForCausalLM": "proxy_qwen2.MobilintQwen2ForCausalLM"
  },
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_position_embeddings": 16384,
  "max_window_layers": 70,
  "model_type": "mobilint-qwen2",
  "mxq_path": "Qwen2.5-3B-Instruct-W4V8.mxq",
  "num_attention_heads": 16,
  "num_hidden_layers": 36,
  "num_key_value_heads": 2,
  "rms_norm_eps": 1e-06,
  "rope_theta": 1000000.0,
  "sliding_window": 32768,
  "target_cores": [
    "0:0"
  ],
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.43.1",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936,
  "npu_prefill_chunk_size": {
    "global4": 512,
    "global8": 1024,
    "single": 512
  },
  "max_batch_size": 1
}