Instructions to use IFM/K2-Horizon-32B-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IFM/K2-Horizon-32B-NVFP4 with Transformers:

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

pipe = pipeline("text-generation", model="IFM/K2-Horizon-32B-NVFP4", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("IFM/K2-Horizon-32B-NVFP4", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use IFM/K2-Horizon-32B-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IFM/K2-Horizon-32B-NVFP4"
# 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-32B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IFM/K2-Horizon-32B-NVFP4

SGLang

How to use IFM/K2-Horizon-32B-NVFP4 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-32B-NVFP4" \
    --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-32B-NVFP4",
		"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-32B-NVFP4" \
        --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-32B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IFM/K2-Horizon-32B-NVFP4 with Docker Model Runner:
```
docker model run hf.co/IFM/K2-Horizon-32B-NVFP4
```

why nvfp4 for stage 1?

by pirola - opened Sep 23

Discussion

pirola

Sep 23

•

edited Sep 23

i would expect an nvfp4 release on a final checkpoint, why on a stage1?
and BTW, when do you plan the final checkpoint release?
I was porting a quantized version of IFM/K2-Horizon-32B to run in FreeToken on a 16Gb VRAM, but just found out the KV cache for this model is GIGANTIC!!! hahaha a 256k 8-bits kv alone would take more than my GPU has to offer. that's sad (for the VRAM-poor)

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