Instructions to use ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Local Apps Settings

How to use ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128

SGLang

How to use ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128 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 "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128" \
    --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": "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128",
		"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 "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128" \
        --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": "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128 with Docker Model Runner:
```
docker model run hf.co/ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128
```

Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128

This is a GPTQ-quantized version of microsoft/Phi-tiny-MoE-instruct, created with GPTQModel 7.3.4. It is intended as a small mixed-group-size PhiMoE checkpoint and loader integration fixture.

What is special about this checkpoint?

The global configuration is symmetric W4G128. Exact per-module dynamic overrides use W4G32 for attention Q/K/V and every expert gate/up projection. This is mixed group-size quantization; all quantized weights use 4 bits.

Projection family	Modules per layer	Configuration	Modules in 32 layers
Attention Q/K/V	3	W4G32	96
Attention output	1	W4G128	32
16 experts x gate/up	32	W4G32	1,024
16 experts x down	16	W4G128	512
Total	52		1,664

Across the complete model:

All 32 decoder layers and all 16 experts per layer are quantized.
1,120 projections use W4G32 dynamic overrides.
544 projections use the global W4G128 setting.
The 32 routers and lm_head remain dense by design.
qweight, qzeros, scales, and g_idx are present for all 1,664 quantized projections.

Quantization details

Setting	Value
Source dtype	BF16
Method / format	GPTQ / `gptq`
Bits	4 everywhere
Global group size	128
Dynamic group size	32
Symmetric	Yes
Activation order (`desc_act`)	No
Group-aware activation ordering	Yes
Sequential quantization	Yes
MSE search	2
Pack dtype / implementation	INT32 / GPU
Calibration samples	257
Calibration concatenation size	4,096 tokens
Observed attention calibration tokens	97,946
Checkpoint layout	One unsharded `model.safetensors` file
Quantization runtime	GPTQModel 7.3.4, Transformers 5.14.1, Torch 2.9.1+cu130

Every expert received a nonzero number of calibration activations during quantization.

Loading

from gptqmodel import BACKEND, GPTQModel

model = GPTQModel.load(
    "ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128",
    backend=BACKEND.GPTQ_TORCH,
    device="cuda",
)

Validation reloads all 1,664 quantized projections with the expected group sizes and completes a finite-logit forward pass. No downstream quality benchmark is claimed; evaluate the checkpoint for the intended use case.

Downloads last month: -

Safetensors

Model size

4B params

Tensor type

BF16

F16

I32

Model tree for ModelCloud/Phi-tiny-MoE-instruct-GPTQ-W4-MixedGroup-G32-G128

Base model

microsoft/Phi-tiny-MoE-instruct

Quantized

(2)

this model