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
```

hunterhector commited on Sep 8, 2025

Commit

6cb377c

verified ·

1 Parent(s): f0e8595

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -30

README.md CHANGED Viewed

@@ -7,24 +7,13 @@ library_name: transformers
 base_model: Qwen/Qwen2.5-32B
 ---
-<p align="center">
-  <img alt="k2-think-banner" src="https://github.com/MBZUAI-IFM/K2-Think-SFT/blob/main/assets/k2-think-banner.png">
-</p>
 <p align="center">
   <a href="https://k2think.ai"><strong>Try K2-Think</strong></a> ·
-  <a href="https://arxiv.org/abs/xxxxxxxx"><strong>Tech Report</strong></a> ·
-  <a href="https://github.com/LLM360/Reasoning360"><strong>Code </strong></a>
 </p>
 <br>
-K2-Think is a 32 billion parameter open-weights general reasoning model with strong performance in competitive mathematical problem solving. Built on a Qwen2.5-32B base, K2-Think combines long CoT SFT, RL with verifiable rewards, and a test-time scaling scaffold to match or exceed much larger models on public math benchmarks while keeping latency low.
-## **Highlights**
-- **Math specialist at 32B:** State-of-the-art results among open models on AIME-style olympiad math and other hard math sets.
-- **Fast generation:** **~2,000 tokens/sec** on our Cerebras WSE deployment; **~10×** faster than typical H100/H200 setups in our tests.
-- **Token-efficient reasoning:** Planning reduces average response length by **up to ~14%** at equal or higher accuracy.
 # Quickstart
@@ -82,24 +71,6 @@ We deploy K2-THINK on Cerebras Wafer-Scale Engine (WSE) systems, leveraging the
 | **Cerebras WSE (our deployment)** |             **\~2,000** |                         **\~16 s** |
 | Typical **H100/H200** GPU setup   |                   \~200 |                            \~160 s |
----
-## Token Efficiency
-K2-Think's **Plan-Before-You-Think** methodology combined with Best-of-N sampling produces more concise reasoning chains while maintaining or improving accuracy. Our test-time scaffold reduces average response length by up to **14%** across mathematical benchmarks.
-**Token reduction** per completed answer (SFT+RL checkpoint vs K2-Think):
-| Domain   | Benchmark      | SFT+RL Checkpoint |   K2-Think |          Δ |
-| -------- | -------------- | ----------------: | ---------: | ---------: |
-| **Math** | AIME24         |            23,324 | **20,058** | **−14.0%** |
-| **Math** | AIME25         |            25,869 | **24,218** | **−6.38%** |
-| **Math** | HMMT25         |            31,475 | **26,977** | **−14.3%** |
-| **Math** | OMNI-Math-HARD |            35,266 | **30,032** | **−14.0%** |
-| Code     | LiveCodeBench  |            13,552 | **12,166** | **−10.2%** |
-| Science  | GPQA-Diamond   |            15,271 | **14,661** | **−3.99%** |
 ---
 ## Safety Evaluation

 base_model: Qwen/Qwen2.5-32B
 ---
 <p align="center">
   <a href="https://k2think.ai"><strong>Try K2-Think</strong></a> ·
 </p>
 <br>
+K2-Think is a 32 billion parameter open-weights general reasoning model with strong performance in competitive mathematical problem solving.
 # Quickstart
 | **Cerebras WSE (our deployment)** |             **\~2,000** |                         **\~16 s** |
 | Typical **H100/H200** GPU setup   |                   \~200 |                            \~160 s |
 ---
 ## Safety Evaluation