Instructions to use jkim96/phi-4-DASHQ-INT3-g32 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jkim96/phi-4-DASHQ-INT3-g32 with Transformers:

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

pipe = pipeline("text-generation", model="jkim96/phi-4-DASHQ-INT3-g32", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jkim96/phi-4-DASHQ-INT3-g32", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("jkim96/phi-4-DASHQ-INT3-g32", trust_remote_code=True, 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 jkim96/phi-4-DASHQ-INT3-g32 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jkim96/phi-4-DASHQ-INT3-g32"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jkim96/phi-4-DASHQ-INT3-g32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jkim96/phi-4-DASHQ-INT3-g32

SGLang

How to use jkim96/phi-4-DASHQ-INT3-g32 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 "jkim96/phi-4-DASHQ-INT3-g32" \
    --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": "jkim96/phi-4-DASHQ-INT3-g32",
		"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 "jkim96/phi-4-DASHQ-INT3-g32" \
        --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": "jkim96/phi-4-DASHQ-INT3-g32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jkim96/phi-4-DASHQ-INT3-g32 with Docker Model Runner:
```
docker model run hf.co/jkim96/phi-4-DASHQ-INT3-g32
```

phi-4-DASHQ-INT3-g32 / README.md

jkim96

Upload DASH-Q quantized checkpoint (INTNone, gNone)

2f79daa verified 4 days ago

preview code

Raw

History Blame

2.26 kB

	---
	license: apache-2.0
	base_model: microsoft/phi-4
	library_name: transformers
	tags:
	- dashq
	- quantized
	- post-training-quantization
	---
	# phi-4-DASHQ-INT3-g32

	This repository contains a DASH-Q packed quantized checkpoint for `microsoft/phi-4`.

	DASH-Q checkpoints require the lightweight DASH-Q runtime package for loading. They are not plain Transformers checkpoints because linear layers are stored as `PackedQuantizedLinear` modules.

	## Usage

	```python
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model = AutoModelForCausalLM.from_pretrained(
	"jkim96/phi-4-DASHQ-INT3-g32", trust_remote_code=True, device_map="cuda", dtype="auto"
	)
	tokenizer = AutoTokenizer.from_pretrained("jkim96/phi-4-DASHQ-INT3-g32")

	messages = [{"role": "user", "content": "Explain 2-bit quantization in one sentence."}]
	text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
	inputs = tokenizer(text, return_tensors="pt").to(model.device)
	print(tokenizer.decode(model.generate(**inputs, max_new_tokens=256)[0]))
	```

	`trust_remote_code=True` is required: the checkpoint ships its quantized-layer
	implementation (`modeling_dashq.py`) and Triton kernels (`dashq_kernel.py`).
	Without Triton, or on CPU, it falls back to dequantize-and-matmul in PyTorch.

	### Requirements

	\| Package \| Minimum \| Verified with \|
	\| --- \| --- \| --- \|
	\| `torch` \| 2.4 \| 2.12.1+cu130 \|
	\| `transformers` \| 5.8 \| 5.9.0 \|
	\| `triton` \| 3.0 (Linux; bundled with CUDA builds of PyTorch) \| 3.7.1 \|
	\| `huggingface_hub` \| 1.5 (pulled in by transformers) \| 1.15.0 \|

	## Quantization

	\| Field \| Value \|
	\| --- \| --- \|
	\| Base model \| `microsoft/phi-4` \|
	\| Bits \| `3` \|
	\| Group size \| `32` \|
	\| Scale/zero dtype \| `float16` \|
	\| Calibration dataset \| `wikitext2` \|
	\| Calibration samples \| `128` \|
	\| Sequence length \| `2048` \|
	\| Original size \| `29.3190 GB` \|
	\| Quantized size \| `9.2126 GB` \|

	## Evaluation

	\| Metric \| Value \|
	\| --- \| ---: \|
	\| `wikitext2_ppl` \| 6.6497 \|
	\| `zero-shot accuracy avg` \| 69.2339 \|
	\| `arc_challenge` \| 57.2526 \|
	\| `arc_easy` \| 74.9158 \|
	\| `commonsense_qa` \| 75.5119 \|
	\| `hellaswag` \| 80.5218 \|
	\| `lambada_openai` \| 73.9375 \|
	\| `openbookqa` \| 45.8000 \|
	\| `piqa` \| 81.1208 \|
	\| `truthfulqa_mc2` \| 59.5376 \|
	\| `winogrande` \| 74.5067 \|

	---
	license: apache-2.0
	base_model: microsoft/phi-4
	library_name: transformers
	tags:
	- dashq
	- quantized
	- post-training-quantization
	---
	# phi-4-DASHQ-INT3-g32

	This repository contains a DASH-Q packed quantized checkpoint for `microsoft/phi-4`.

	DASH-Q checkpoints require the lightweight DASH-Q runtime package for loading. They are not plain Transformers checkpoints because linear layers are stored as `PackedQuantizedLinear` modules.

	## Usage

	```python
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model = AutoModelForCausalLM.from_pretrained(
	"jkim96/phi-4-DASHQ-INT3-g32", trust_remote_code=True, device_map="cuda", dtype="auto"
	)
	tokenizer = AutoTokenizer.from_pretrained("jkim96/phi-4-DASHQ-INT3-g32")

	messages = [{"role": "user", "content": "Explain 2-bit quantization in one sentence."}]
	text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
	inputs = tokenizer(text, return_tensors="pt").to(model.device)
	print(tokenizer.decode(model.generate(**inputs, max_new_tokens=256)[0]))
	```

	`trust_remote_code=True` is required: the checkpoint ships its quantized-layer
	implementation (`modeling_dashq.py`) and Triton kernels (`dashq_kernel.py`).
	Without Triton, or on CPU, it falls back to dequantize-and-matmul in PyTorch.

	### Requirements

	\| Package \| Minimum \| Verified with \|
	\| --- \| --- \| --- \|
	\| `torch` \| 2.4 \| 2.12.1+cu130 \|
	\| `transformers` \| 5.8 \| 5.9.0 \|
	\| `triton` \| 3.0 (Linux; bundled with CUDA builds of PyTorch) \| 3.7.1 \|
	\| `huggingface_hub` \| 1.5 (pulled in by transformers) \| 1.15.0 \|

	## Quantization

	\| Field \| Value \|
	\| --- \| --- \|
	\| Base model \| `microsoft/phi-4` \|
	\| Bits \| `3` \|
	\| Group size \| `32` \|
	\| Scale/zero dtype \| `float16` \|
	\| Calibration dataset \| `wikitext2` \|
	\| Calibration samples \| `128` \|
	\| Sequence length \| `2048` \|
	\| Original size \| `29.3190 GB` \|
	\| Quantized size \| `9.2126 GB` \|

	## Evaluation

	\| Metric \| Value \|
	\| --- \| ---: \|
	\| `wikitext2_ppl` \| 6.6497 \|
	\| `zero-shot accuracy avg` \| 69.2339 \|
	\| `arc_challenge` \| 57.2526 \|
	\| `arc_easy` \| 74.9158 \|
	\| `commonsense_qa` \| 75.5119 \|
	\| `hellaswag` \| 80.5218 \|
	\| `lambada_openai` \| 73.9375 \|
	\| `openbookqa` \| 45.8000 \|
	\| `piqa` \| 81.1208 \|
	\| `truthfulqa_mc2` \| 59.5376 \|
	\| `winogrande` \| 74.5067 \|