Instructions to use inference-optimization/Qwen3-1.6B-A0.9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use inference-optimization/Qwen3-1.6B-A0.9B with Transformers:

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

pipe = pipeline("text-generation", model="inference-optimization/Qwen3-1.6B-A0.9B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("inference-optimization/Qwen3-1.6B-A0.9B")
model = AutoModelForCausalLM.from_pretrained("inference-optimization/Qwen3-1.6B-A0.9B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use inference-optimization/Qwen3-1.6B-A0.9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "inference-optimization/Qwen3-1.6B-A0.9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "inference-optimization/Qwen3-1.6B-A0.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/inference-optimization/Qwen3-1.6B-A0.9B

SGLang

How to use inference-optimization/Qwen3-1.6B-A0.9B 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 "inference-optimization/Qwen3-1.6B-A0.9B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "inference-optimization/Qwen3-1.6B-A0.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "inference-optimization/Qwen3-1.6B-A0.9B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "inference-optimization/Qwen3-1.6B-A0.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use inference-optimization/Qwen3-1.6B-A0.9B with Docker Model Runner:
```
docker model run hf.co/inference-optimization/Qwen3-1.6B-A0.9B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Qwen3-1.6B-A0.9B

This is a tiny version of Qwen/Qwen3-30B-A3B created for testing and development.

Model Details

Base Model: Qwen/Qwen3-30B-A3B
Architecture: qwen3_moe (Mixture of Experts)
Total Parameters: 1.57B
Activated Parameters: ~0.9B (50% MoE activation)

Configuration Changes

The following parameters were reduced from the original model:

Parameter	Original	Tiny
num_hidden_layers	48	10
num_local_experts	128	16
num_experts_per_tok	8	8
hidden_size	2048	2048
intermediate_size	6144	6144
moe_intermediate_size	768	768
num_attention_heads	32	32
num_key_value_heads	4	4

Checkpoint Structure

The checkpoint is stored as a single model.safetensors file with individual expert weights matching the original Qwen3 structure. Each layer has 16 experts with separate gate_proj, up_proj, and down_proj weights per expert.

Validation

The model was fine-tuned on a toy copypasta dataset and achieves:

Perplexity: 1.0 (on validation text)
Generation: Successfully generates coherent continuations

Example generation:

Input: "According to all known laws"
Output: "According to all known laws of aviation, there is no way a bee should be able to fly."

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("inference-optimization/Qwen3-1.6B-A0.9B", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("inference-optimization/Qwen3-1.6B-A0.9B")

input_ids = tokenizer("According to all known laws", return_tensors="pt").input_ids.to(model.device)
output = model.generate(input_ids, max_new_tokens=20)
print(tokenizer.decode(output[0]))

Creation Process

This model was created using the llm-compressor create-tiny-model Claude skill:

Configuration: Created with 10 layers and 16 experts (8 activated per token) to achieve ~1.6B total parameters with 50% MoE activation
Initialization: Randomly initialized weights using transformers init_weights()
Fine-tuning: Trained on famous internet copypastas until perplexity < 3.0
Checkpoint Conversion: Converted from batched expert format to individual expert format to match original Qwen3 checkpoint structure
Validation: Confirmed perplexity ~1.0 and successful text generation

Notes

This model maintains the same MoE architecture as the original with Grouped Query Attention (GQA)
The checkpoint format exactly matches the original Qwen3-30B-A3B structure for compatibility
This model is intended for testing and development only, not production use

Downloads last month: 114

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for inference-optimization/Qwen3-1.6B-A0.9B

Base model

Qwen/Qwen3-30B-A3B-Base

Finetuned

Qwen/Qwen3-30B-A3B

Finetuned

(61)

this model

Quantizations

2 models