Instructions to use RadixArk/Kimi-K3-DSpark with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RadixArk/Kimi-K3-DSpark with Transformers:

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

pipe = pipeline("text-generation", model="RadixArk/Kimi-K3-DSpark", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("RadixArk/Kimi-K3-DSpark", trust_remote_code=True)
model = AutoModel.from_pretrained("RadixArk/Kimi-K3-DSpark", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RadixArk/Kimi-K3-DSpark with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RadixArk/Kimi-K3-DSpark"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RadixArk/Kimi-K3-DSpark",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RadixArk/Kimi-K3-DSpark

SGLang

How to use RadixArk/Kimi-K3-DSpark 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/Kimi-K3-DSpark" \
    --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/Kimi-K3-DSpark",
		"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/Kimi-K3-DSpark" \
        --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/Kimi-K3-DSpark",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RadixArk/Kimi-K3-DSpark with Docker Model Runner:
```
docker model run hf.co/RadixArk/Kimi-K3-DSpark
```

Training compute

by kshou-kai - opened 1 day ago

Discussion

kshou-kai

1 day ago

Could you share the approximate GPU-hours required for (1) regenerating the training data with Kimi-K3 and (2) training the DSpark draft model? Even rough numbers would be very helpful.

skx618

RadixArk org 1 day ago

Consider 16 GB300s. The cost of regeneration depends on the source dataset and generation configuration. In general, generating sufficient training data is expected to take approximately one week. Training is anticipated to take 1–3 days, depending on the readiness and volume of the regenerated data.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment