Instructions to use IFM/K2-Horizon-0.9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IFM/K2-Horizon-0.9B with Transformers:

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

pipe = pipeline("text-generation", model="IFM/K2-Horizon-0.9B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("IFM/K2-Horizon-0.9B", trust_remote_code=True, device_map="auto")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use IFM/K2-Horizon-0.9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IFM/K2-Horizon-0.9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IFM/K2-Horizon-0.9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IFM/K2-Horizon-0.9B

SGLang

How to use IFM/K2-Horizon-0.9B 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 "IFM/K2-Horizon-0.9B" \
    --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": "IFM/K2-Horizon-0.9B",
		"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 "IFM/K2-Horizon-0.9B" \
        --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": "IFM/K2-Horizon-0.9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IFM/K2-Horizon-0.9B with Docker Model Runner:
```
docker model run hf.co/IFM/K2-Horizon-0.9B
```

K2-Horizon-0.9B License Clarification

by Vo-Tron - opened 5 days ago

Discussion

Vo-Tron

5 days ago

Hello,

I have a question about the licensing/usage terms for IFM/K2-Horizon-0.9B.

The Hugging Face repository currently identifies the model as Apache-2.0, but I noticed residual metadata associated with the model that appears to indicate internal-only.

Could you please clarify which status is authoritative?

Specifically:

Are the internal-only fields intentional and currently applicable to K2-Horizon-0.9B, or are they stale/incorrect metadata?
If they are stale, is there an authoritative revision or repository update that corrects them?
Is Apache-2.0 the governing license for the released model and weights?
Does that license/status permit:
- research and evaluation;
- local inference/use;
- fine-tuning or other derivative work;
- redistribution;
- commercial use?

If there is a specific LICENSE file, commit, or first-party document that governs these rights, a link to it would be greatly appreciated.

Thank you.

hunterhector

Institute of Foundation Models org 5 days ago

Hi, @Vo-Tron

For your questions:

No, there are no intentional internal-only field from our side. Could you also help point out what that is? We can investigate and correct the metadata.
We will if we can identify them.
Yes it is Apache-2.0
Yes that's the implication of the license and our goal for open source
We will check. If we miss the upload we will do so.

Vo-Tron

5 days ago

Thank you. The fields I was referring to are in the README front matter:

license_name: internal-only
license_link: LICENSE

I also observed the same licensing pattern on the -0.9B-GGUF repository.

That appears inconsistent with the repository's Apache-2.0 designation and your clarification above. I wanted to point you to the exact fields you asked about.

Thank you for checking.

hunterhector

Institute of Foundation Models org 5 days ago

This is actually a mistake when we batch processed our documents. We will fix these.

Thanks for catching this!

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