Instructions to use AlexCuadron/Qwen-32B-8a4e8f3a with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AlexCuadron/Qwen-32B-8a4e8f3a with Transformers:

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

pipe = pipeline("text-generation", model="AlexCuadron/Qwen-32B-8a4e8f3a")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AlexCuadron/Qwen-32B-8a4e8f3a")
model = AutoModelForCausalLM.from_pretrained("AlexCuadron/Qwen-32B-8a4e8f3a", 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 AlexCuadron/Qwen-32B-8a4e8f3a with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AlexCuadron/Qwen-32B-8a4e8f3a"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AlexCuadron/Qwen-32B-8a4e8f3a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AlexCuadron/Qwen-32B-8a4e8f3a

SGLang

How to use AlexCuadron/Qwen-32B-8a4e8f3a 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 "AlexCuadron/Qwen-32B-8a4e8f3a" \
    --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": "AlexCuadron/Qwen-32B-8a4e8f3a",
		"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 "AlexCuadron/Qwen-32B-8a4e8f3a" \
        --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": "AlexCuadron/Qwen-32B-8a4e8f3a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AlexCuadron/Qwen-32B-8a4e8f3a with Docker Model Runner:
```
docker model run hf.co/AlexCuadron/Qwen-32B-8a4e8f3a
```

Qwen-32B-8a4e8f3a

File size: 2,072 Bytes

031fedf

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9696969696969697,
  "eval_steps": 500,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.50050950050354,
      "learning_rate": 2.5e-06,
      "loss": 0.8407,
      "step": 1
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.2383944988250732,
      "learning_rate": 5e-06,
      "loss": 0.7878,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.1805531978607178,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.8026,
      "step": 3
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.0964150428771973,
      "learning_rate": 1e-05,
      "loss": 0.8133,
      "step": 4
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 1.5487000942230225,
      "learning_rate": 9.968561049466214e-06,
      "loss": 0.7608,
      "step": 5
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.7998461723327637,
      "learning_rate": 9.874639560909118e-06,
      "loss": 0.7272,
      "step": 6
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 1.9523899555206299,
      "learning_rate": 9.719416651541839e-06,
      "loss": 0.6923,
      "step": 7
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 4.303847312927246,
      "learning_rate": 9.504844339512096e-06,
      "loss": 0.6853,
      "step": 8
    }
  ],
  "logging_steps": 1,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 21013697134592.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}