Instructions to use RadixArk/Inkling-DSpark-Preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RadixArk/Inkling-DSpark-Preview with Transformers:

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

pipe = pipeline("text-generation", model="RadixArk/Inkling-DSpark-Preview", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("RadixArk/Inkling-DSpark-Preview", trust_remote_code=True)
model = AutoModel.from_pretrained("RadixArk/Inkling-DSpark-Preview", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RadixArk/Inkling-DSpark-Preview with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/RadixArk/Inkling-DSpark-Preview

SGLang

How to use RadixArk/Inkling-DSpark-Preview 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 "RadixArk/Inkling-DSpark-Preview" \
    --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": "RadixArk/Inkling-DSpark-Preview",
		"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 "RadixArk/Inkling-DSpark-Preview" \
        --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": "RadixArk/Inkling-DSpark-Preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RadixArk/Inkling-DSpark-Preview with Docker Model Runner:
```
docker model run hf.co/RadixArk/Inkling-DSpark-Preview
```

Inkling-DSpark-Preview / config.json

cm00cm

final checkpoint: epoch_10_step_311200

21e29a6 verified 16 days ago

Raw

History Blame Contribute Delete

1.33 kB

	{
	"architectures": [
	"DSparkDraftModel"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoModel": "dspark.DSparkDraftModel"
	},
	"block_size": 7,
	"bos_token_id": null,
	"confidence_head_with_markov": true,
	"dflash_config": {
	"mask_token_id": 200064,
	"target_layer_ids": [
	5,
	17,
	35,
	47,
	59
	]
	},
	"dtype": "bfloat16",
	"enable_confidence_head": true,
	"eos_token_id": [
	200006,
	200000,
	200002,
	200003
	],
	"head_dim": 64,
	"hidden_act": "silu",
	"hidden_size": 6144,
	"initializer_range": 0.02,
	"intermediate_size": 12288,
	"layer_types": [
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention"
	],
	"markov_head_type": "vanilla",
	"markov_rank": 256,
	"max_position_embeddings": 1048576,
	"max_window_layers": 5,
	"model_type": "qwen3",
	"num_attention_heads": 64,
	"num_hidden_layers": 5,
	"num_key_value_heads": 16,
	"num_target_layers": 66,
	"pad_token_id": 200006,
	"rms_norm_eps": 1e-05,
	"rope_parameters": {
	"rope_theta": 8000000,
	"rope_type": "default"
	},
	"sliding_window": null,
	"tie_word_embeddings": false,
	"transformers_version": "5.8.1",
	"use_cache": true,
	"use_sliding_window": false,
	"vocab_size": 201024
	}