Instructions to use Adityapande403/SpecDecode-edge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Adityapande403/SpecDecode-edge with Transformers:

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

pipe = pipeline("text-generation", model="Adityapande403/SpecDecode-edge")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Adityapande403/SpecDecode-edge", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Adityapande403/SpecDecode-edge with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Adityapande403/SpecDecode-edge"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Adityapande403/SpecDecode-edge",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Adityapande403/SpecDecode-edge

SGLang

How to use Adityapande403/SpecDecode-edge 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 "Adityapande403/SpecDecode-edge" \
    --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": "Adityapande403/SpecDecode-edge",
		"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 "Adityapande403/SpecDecode-edge" \
        --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": "Adityapande403/SpecDecode-edge",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Adityapande403/SpecDecode-edge with Docker Model Runner:
```
docker model run hf.co/Adityapande403/SpecDecode-edge
```

SpecDecode-edge / config.json

Adityapande403

Upload Medusa-3Heads speculative weights and config for Qwen2.5-1.5B-Instruct

a1ad09c verified 2 days ago

Raw History Blame Contribute Delete

308 Bytes

{
  "model_type": "medusa",
  "base_model_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
  "num_heads": 3,
  "num_layers": 1,
  "hidden_size": 1536,
  "vocab_size": 151936,
  "torch_dtype": "bfloat16",
  "architectures": [
    "MedusaModel"
  ],
  "medusa_num_heads": 3,
  "medusa_num_layers": 1
}

	{
	"model_type": "medusa",
	"base_model_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
	"num_heads": 3,
	"num_layers": 1,
	"hidden_size": 1536,
	"vocab_size": 151936,
	"torch_dtype": "bfloat16",
	"architectures": [
	"MedusaModel"
	],
	"medusa_num_heads": 3,
	"medusa_num_layers": 1
	}