Instructions to use LLM360/K2-Think with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM360/K2-Think with Transformers:

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

pipe = pipeline("text-generation", model="LLM360/K2-Think")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LLM360/K2-Think")
model = AutoModelForCausalLM.from_pretrained("LLM360/K2-Think")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use LLM360/K2-Think with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM360/K2-Think"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM360/K2-Think",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LLM360/K2-Think

SGLang

How to use LLM360/K2-Think 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 "LLM360/K2-Think" \
    --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": "LLM360/K2-Think",
		"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 "LLM360/K2-Think" \
        --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": "LLM360/K2-Think",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LLM360/K2-Think with Docker Model Runner:
```
docker model run hf.co/LLM360/K2-Think
```

nielsr HF Staff commited on Sep 10, 2025

Commit

46b668c

verified ·

1 Parent(s): af3cc43

Improve model card: Add paper, code, and project page links, and main title

Browse files

This PR enhances the model card by:

- Adding the paper title as a prominent H1 heading for better structure.
- Adding direct links to the paper ([https://huggingface.co/papers/2509.07604](https://huggingface.co/papers/2509.07604)), the GitHub repository ([https://github.com/MBZUAI-IFM/K2-Think-SFT](https://github.com/MBZUAI-IFM/K2-Think-SFT)), and the project page ([https://k2think.ai](https://k2think.ai)) at the top of the model card. This makes essential resources more discoverable for users.
- Ensuring proper Markdown rendering for special characters such as `\~` in the "Inference Speed" table by escaping them as `\\~`.

These updates improve the model card's clarity, navigability, and adherence to best practices for documenting artifacts on the Hub.

Files changed (1) hide show

README.md +9 -4

README.md CHANGED Viewed

@@ -1,11 +1,16 @@
 ---
 language:
 - en
 license: apache-2.0
 pipeline_tag: text-generation
-library_name: transformers
-base_model: Qwen/Qwen2.5-32B
 ---
 <center><img src="banner.png" alt="k2-think-banner"/></center>
 <p align="center">
@@ -69,8 +74,8 @@ We deploy K2-THINK on Cerebras Wafer-Scale Engine (WSE) systems, leveraging the
 | Platform                          | Throughput (tokens/sec) | Example: 32k-token response (time) |
 | --------------------------------- | ----------------------: | ---------------------------------: |
-| **Cerebras WSE (our deployment)** |             **\~2,000** |                         **\~16 s** |
-| Typical **H100/H200** GPU setup   |                   \~200 |                            \~160 s |
 ---

 ---
+base_model: Qwen/Qwen2.5-32B
 language:
 - en
+library_name: transformers
 license: apache-2.0
 pipeline_tag: text-generation
 ---
+# K2-Think: A Parameter-Efficient Reasoning System
+📚 [Paper](https://huggingface.co/papers/2509.07604) - 📝 [Code](https://github.com/MBZUAI-IFM/K2-Think-SFT) - 🏢 [Project Page](https://k2think.ai)
 <center><img src="banner.png" alt="k2-think-banner"/></center>
 <p align="center">
 | Platform                          | Throughput (tokens/sec) | Example: 32k-token response (time) |
 | --------------------------------- | ----------------------: | ---------------------------------: |
+| **Cerebras WSE (our deployment)** |             **\\~2,000** |                         **\\~16 s** |
+| Typical **H100/H200** GPU setup   |                   \\~200 |                            \\~160 s |
 ---