Instructions to use wassname/qwen3-5lyr-tiny-random with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use wassname/qwen3-5lyr-tiny-random with Transformers:

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

pipe = pipeline("text-generation", model="wassname/qwen3-5lyr-tiny-random")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("wassname/qwen3-5lyr-tiny-random")
model = AutoModelForCausalLM.from_pretrained("wassname/qwen3-5lyr-tiny-random", 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 wassname/qwen3-5lyr-tiny-random with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "wassname/qwen3-5lyr-tiny-random"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "wassname/qwen3-5lyr-tiny-random",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/wassname/qwen3-5lyr-tiny-random

SGLang

How to use wassname/qwen3-5lyr-tiny-random 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 "wassname/qwen3-5lyr-tiny-random" \
    --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": "wassname/qwen3-5lyr-tiny-random",
		"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 "wassname/qwen3-5lyr-tiny-random" \
        --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": "wassname/qwen3-5lyr-tiny-random",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use wassname/qwen3-5lyr-tiny-random with Docker Model Runner:
```
docker model run hf.co/wassname/qwen3-5lyr-tiny-random
```

Model Card for Model ID

Code used to create this, 5 layer version of https://huggingface.co/tiny-random/qwen3

This tiny model is for debugging. It is randomly initialized with the config adapted from Qwen/Qwen3-32B.

import torch

from transformers import (
    AutoConfig,
    AutoModelForCausalLM,
    AutoTokenizer,
    GenerationConfig,
    pipeline,
    set_seed,
)

source_model_id = "Qwen/Qwen3-32B"
save_folder = "/tmp/tiny-random/qwen3-5lyr"

tokenizer = AutoTokenizer.from_pretrained(
    source_model_id, trust_remote_code=True,
)
tokenizer.save_pretrained(save_folder)

config = AutoConfig.from_pretrained(
    source_model_id, trust_remote_code=True,
)
config._name_or_path = source_model_id
config.hidden_size = 64
config.intermediate_size = 128
config.head_dim = 32
config.num_key_value_heads = 1
config.num_attention_heads = 2
config.num_hidden_layers = 5 # modified from https://huggingface.co/tiny-random/qwen3
config.max_window_layers = 1
config.tie_word_embeddings = True
model = AutoModelForCausalLM.from_config(
    config,
    torch_dtype=torch.bfloat16,
    trust_remote_code=True,
)
model.generation_config = GenerationConfig.from_pretrained(
    source_model_id, trust_remote_code=True,
)
set_seed(42)
with torch.no_grad():
    for name, p in sorted(model.named_parameters()):
        torch.nn.init.normal_(p, 0, 0.5)
        print(name, p.shape)
model.save_pretrained(save_folder)

Downloads last month: 1,991

Safetensors

Model size

9.91M params

Tensor type

BF16