Instructions to use moonshotai/Kimi-K2.6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moonshotai/Kimi-K2.6 with Transformers:

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

pipe = pipeline("image-text-to-text", model="moonshotai/Kimi-K2.6", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModel

processor = AutoProcessor.from_pretrained("moonshotai/Kimi-K2.6", trust_remote_code=True)
model = AutoModel.from_pretrained("moonshotai/Kimi-K2.6", trust_remote_code=True, device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use moonshotai/Kimi-K2.6 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moonshotai/Kimi-K2.6"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moonshotai/Kimi-K2.6",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/moonshotai/Kimi-K2.6

SGLang

How to use moonshotai/Kimi-K2.6 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 "moonshotai/Kimi-K2.6" \
    --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": "moonshotai/Kimi-K2.6",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "moonshotai/Kimi-K2.6" \
        --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": "moonshotai/Kimi-K2.6",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use moonshotai/Kimi-K2.6 with Docker Model Runner:
```
docker model run hf.co/moonshotai/Kimi-K2.6
```

JustinTong commited on Apr 20

Commit

b5632fb

verified ·

1 Parent(s): 2b2b88e

Update SGLang deployment guide: point to K2.6 cookbook and use stable release

Browse files

The SGLang section previously linked to the Kimi-K2.5 cookbook and required a `git+https://github.com/sgl-project/sglang.git` install from `main`. Kimi-K2.6 now has a dedicated cookbook page and is supported in the SGLang v0.5.10 stable release (available on PyPI), so a nightly/main build is no longer required.

Changes (SGLang section only):
- Update cookbook URL: Kimi-K2.5 -> Kimi-K2.6 (https://cookbook.sglang.io/autoregressive/Moonshotai/Kimi-K2.6)
- Replace "latest main" git install with `pip install "sglang[all]>=0.5.10"`
- Drop the separate `nvidia-cudnn-cu12==9.16.0.29` pin (bundled with the stable wheel)

No changes to vLLM or KTransformers sections.

Files changed (1) hide show

docs/deploy_guidance.md +4 -5

docs/deploy_guidance.md CHANGED Viewed

@@ -28,16 +28,15 @@ vllm serve $MODEL_PATH -tp 8 --mm-encoder-tp-mode data --trust-remote-code --too
 ## SGLang Deployment
-You can refer to https://cookbook.sglang.io/autoregressive/Moonshotai/Kimi-K2.5 for the newest deployment guide.
-This model is available in SGLang latest main:
 ```
-pip install "sglang @ git+https://github.com/sgl-project/sglang.git#subdirectory=python"
-pip install nvidia-cudnn-cu12==9.16.0.29
 ```
-Similarly, here is the example for it to run with TP8 on H200 in a single node via SGLang:
 ``` bash
 sglang serve --model-path $MODEL_PATH --tp 8 --trust-remote-code --tool-call-parser kimi_k2 --reasoning-parser kimi_k2
 ```

 ## SGLang Deployment
+You can refer to https://cookbook.sglang.io/autoregressive/Moonshotai/Kimi-K2.6 for the newest deployment guide.
+This model is supported in SGLang v0.5.10 and later stable releases (no nightly / main build required):
 ```
+pip install "sglang[all]>=0.5.10"
 ```
+Here is the example for it to run with TP8 on H200 in a single node via SGLang:
 ``` bash
 sglang serve --model-path $MODEL_PATH --tp 8 --trust-remote-code --tool-call-parser kimi_k2 --reasoning-parser kimi_k2
 ```