Instructions to use amd/tiny-qwen3-moe-w4a8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use amd/tiny-qwen3-moe-w4a8 with Transformers:

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

pipe = pipeline("text-generation", model="amd/tiny-qwen3-moe-w4a8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("amd/tiny-qwen3-moe-w4a8")
model = AutoModelForCausalLM.from_pretrained("amd/tiny-qwen3-moe-w4a8", 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 amd/tiny-qwen3-moe-w4a8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amd/tiny-qwen3-moe-w4a8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amd/tiny-qwen3-moe-w4a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/amd/tiny-qwen3-moe-w4a8

SGLang

How to use amd/tiny-qwen3-moe-w4a8 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 "amd/tiny-qwen3-moe-w4a8" \
    --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": "amd/tiny-qwen3-moe-w4a8",
		"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 "amd/tiny-qwen3-moe-w4a8" \
        --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": "amd/tiny-qwen3-moe-w4a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use amd/tiny-qwen3-moe-w4a8 with Docker Model Runner:
```
docker model run hf.co/amd/tiny-qwen3-moe-w4a8
```

tiny-qwen3-moe-w4a8

File size: 3,339 Bytes

{
  "architectures": [
    "Qwen3MoeForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "decoder_sparse_step": 1,
  "dtype": "bfloat16",
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 2048,
  "max_position_embeddings": 2048,
  "mlp_only_layers": [],
  "model_type": "qwen3_moe",
  "moe_intermediate_size": 1024,
  "norm_topk_prob": false,
  "num_attention_heads": 16,
  "num_experts": 8,
  "num_experts_per_tok": 2,
  "num_hidden_layers": 2,
  "num_key_value_heads": 2,
  "output_router_logits": false,
  "quantization_config": {
    "algo_config": null,
    "exclude": [
      "model.layers.0.self_attn.q_proj",
      "model.layers.0.self_attn.k_proj",
      "model.layers.0.self_attn.v_proj",
      "model.layers.0.self_attn.o_proj",
      "model.layers.0.mlp.gate",
      "model.layers.1.self_attn.q_proj",
      "model.layers.1.self_attn.k_proj",
      "model.layers.1.self_attn.v_proj",
      "model.layers.1.self_attn.o_proj",
      "model.layers.1.mlp.gate",
      "lm_head"
    ],
    "export": {
      "kv_cache_group": [],
      "min_kv_scale": 0.0,
      "pack_method": "reorder",
      "weight_format": "real_quantized",
      "weight_merge_groups": null
    },
    "global_quant_config": {
      "bias": null,
      "input_tensors": {
        "ch_axis": null,
        "dtype": "fp8_e4m3",
        "group_size": null,
        "is_dynamic": true,
        "is_scale_quant": false,
        "mx_element_dtype": null,
        "observer_cls": "PerTensorMinMaxObserver",
        "qscheme": "per_tensor",
        "round_method": "half_even",
        "scale_calculation_mode": null,
        "scale_format": null,
        "scale_type": "float",
        "symmetric": true
      },
      "output_tensors": null,
      "target_device": null,
      "weight": [
        {
          "ch_axis": null,
          "dtype": "fp8_e4m3",
          "group_size": null,
          "is_dynamic": false,
          "is_scale_quant": false,
          "mx_element_dtype": null,
          "observer_cls": "PerTensorMinMaxObserver",
          "qscheme": "per_tensor",
          "round_method": "half_even",
          "scale_calculation_mode": null,
          "scale_format": null,
          "scale_type": "float",
          "symmetric": true
        },
        {
          "ch_axis": 0,
          "dtype": "int4",
          "group_size": null,
          "is_dynamic": false,
          "is_scale_quant": false,
          "mx_element_dtype": null,
          "observer_cls": "PerChannelMinMaxObserver",
          "qscheme": "per_channel",
          "round_method": "half_even",
          "scale_calculation_mode": null,
          "scale_format": null,
          "scale_type": "float",
          "symmetric": true
        }
      ]
    },
    "kv_cache_post_rope": false,
    "kv_cache_quant_config": {},
    "layer_quant_config": {},
    "layer_type_quant_config": {},
    "quant_method": "quark",
    "quant_mode": "eager_mode",
    "softmax_quant_spec": null,
    "version": "0.11.2"
  },
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "router_aux_loss_coef": 0.001,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}