Instructions to use moonshotai/Kimi-K2.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moonshotai/Kimi-K2.5 with Transformers:

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

pipe = pipeline("image-text-to-text", model="moonshotai/Kimi-K2.5", 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 AutoModel
model = AutoModel.from_pretrained("moonshotai/Kimi-K2.5", trust_remote_code=True, dtype="auto")

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use moonshotai/Kimi-K2.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moonshotai/Kimi-K2.5"
# 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-K2.5",
		"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-K2.5

SGLang

How to use moonshotai/Kimi-K2.5 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-K2.5" \
    --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-K2.5",
		"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-K2.5" \
        --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-K2.5",
		"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-K2.5 with Docker Model Runner:
```
docker model run hf.co/moonshotai/Kimi-K2.5
```

400 Bad Request when deploying Kimi-K2.5 with vllm-0.15.0

#60

by tuo02 - opened Feb 5

Discussion

tuo02

Feb 5

Different errors occured when Depoying Kimi-K2.5 with vllm-0.15.0 , while it is no problem when these requests are sended to other models. For example:

Error 1 : ValueError: list index out of range
'''
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] An error occurred in transformers while applying chat template
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] Traceback (most recent call last):
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] File "/usr/local/lib/python3.12/dist-packages/vllm/renderers/hf.py", line 471, in safe_apply_chat_template
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] return tokenizer.apply_chat_template(
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] File "/root/.cache/huggingface/modules/transformers_modules/Kimi_hyphen_K2_dot_5/tokenization_kimi.py", line 338, in apply_chat_template
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] return super().apply_chat_template(
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] File "/usr/local/lib/python3.12/dist-packages/transformers/tokenization_utils_base.py", line 1650, in apply_chat_template
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] isinstance(conversation[0], (list, tuple)) or hasattr(conversation[0], "messages")
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] ~~~~~~~~~~~~^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 05:21:50 [hf.py:483] IndexError: list index out of range
'''

Error 2 : [ValueError: list index out of range](json.decoder.JSONDecodeError: Unterminated string starting at: line 1 column 87 (char 86))
'''
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] Error in preprocessing prompt inputs
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] Traceback (most recent call last):
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/chat_completion/serving.py", line 298, in render_chat_request
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] conversation, engine_prompts = await self._preprocess_chat(
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/engine/serving.py", line 1053, in _preprocess_chat
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] conversation, prompt = await renderer.render_messages_async(
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/local/lib/python3.12/dist-packages/vllm/renderers/hf.py", line 693, in render_messages_async
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] conversation, mm_data, mm_uuids = await parse_chat_messages_async(
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/chat_utils.py", line 1564, in parse_chat_messages_async
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] _postprocess_messages(conversation)
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/chat_utils.py", line 1500, in _postprocess_messages
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] item["function"]["arguments"] = json.loads(content)
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/lib/python3.12/json/init.py", line 346, in loads
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] return _default_decoder.decode(s)
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/lib/python3.12/json/decoder.py", line 338, in decode
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] obj, end = self.raw_decode(s, idx=_w(s, 0).end())
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] File "/usr/lib/python3.12/json/decoder.py", line 354, in raw_decode
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] obj, end = self.scan_once(s, idx)
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] ^^^^^^^^^^^^^^^^^^^^^^
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:08 [serving.py:314] json.decoder.JSONDecodeError: Unterminated string starting at: line 1 column 87 (char 86)
[0;36m(APIServer pid=1)[0;0m INFO: 172.16.221.200:36198 - "POST /v1/chat/completions HTTP/1.1" 400 Bad Request
[0;36m(APIServer pid=1)[0;0m ERROR 02-02 09:27:11 [serving.py:314] Error in preprocessing prompt inputs
'''

chenxi-kalorona-huang

Apr 2

It reproduces on my end as well (vLLM v0.18.0).

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