Instructions to use fxmarty/tiny-gemma-onnx-quantized-trt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fxmarty/tiny-gemma-onnx-quantized-trt with Transformers:

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

pipe = pipeline("text-generation", model="fxmarty/tiny-gemma-onnx-quantized-trt")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fxmarty/tiny-gemma-onnx-quantized-trt")
model = AutoModelForCausalLM.from_pretrained("fxmarty/tiny-gemma-onnx-quantized-trt", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fxmarty/tiny-gemma-onnx-quantized-trt with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fxmarty/tiny-gemma-onnx-quantized-trt"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fxmarty/tiny-gemma-onnx-quantized-trt",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/fxmarty/tiny-gemma-onnx-quantized-trt

SGLang

How to use fxmarty/tiny-gemma-onnx-quantized-trt 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 "fxmarty/tiny-gemma-onnx-quantized-trt" \
    --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": "fxmarty/tiny-gemma-onnx-quantized-trt",
		"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 "fxmarty/tiny-gemma-onnx-quantized-trt" \
        --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": "fxmarty/tiny-gemma-onnx-quantized-trt",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fxmarty/tiny-gemma-onnx-quantized-trt with Docker Model Runner:
```
docker model run hf.co/fxmarty/tiny-gemma-onnx-quantized-trt
```

tiny-gemma-onnx-quantized-trt / ort_config.json

fxmarty

Upload folder using huggingface_hub

1c34dab verified over 2 years ago

Raw History Blame Contribute Delete

1.02 kB

	{
	"one_external_file": true,
	"opset": null,
	"optimization": {},
	"optimum_version": "1.18.0.dev0",
	"quantization": {
	"activations_dtype": "QInt8",
	"activations_symmetric": true,
	"format": "QDQ",
	"is_static": true,
	"mode": "QLinearOps",
	"nodes_to_exclude": [],
	"nodes_to_quantize": [],
	"operators_to_quantize": [
	"Conv",
	"ConvTranspose",
	"Gemm",
	"Clip",
	"Relu",
	"Reshape",
	"Transpose",
	"Squeeze",
	"Unsqueeze",
	"Resize",
	"MaxPool",
	"AveragePool",
	"MatMul",
	"Split",
	"Gather",
	"Where",
	"InstanceNormalization",
	"LayerNormalization"
	],
	"per_channel": false,
	"qdq_add_pair_to_weight": true,
	"qdq_dedicated_pair": true,
	"qdq_op_type_per_channel_support_to_axis": {
	"MatMul": 1
	},
	"reduce_range": false,
	"weights_dtype": "QInt8",
	"weights_symmetric": true
	},
	"transformers_version": "4.39.0.dev0",
	"use_external_data_format": true
	}