Instructions to use moonshotai/Kimi-K3.1-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moonshotai/Kimi-K3.1-Pro with Transformers:

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

pipe = pipeline("image-text-to-text", model="moonshotai/Kimi-K3.1-Pro", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("moonshotai/Kimi-K3.1-Pro", trust_remote_code=True, device_map="auto")

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use moonshotai/Kimi-K3.1-Pro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moonshotai/Kimi-K3.1-Pro"
# 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-K3.1-Pro",
		"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-K3.1-Pro

SGLang

How to use moonshotai/Kimi-K3.1-Pro 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-K3.1-Pro" \
    --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-K3.1-Pro",
		"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-K3.1-Pro" \
        --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-K3.1-Pro",
		"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-K3.1-Pro with Docker Model Runner:
```
docker model run hf.co/moonshotai/Kimi-K3.1-Pro
```

Great release! Please make also small ones. like 27B/A3B or 9B or something like that

by ZeroWw - opened 2 days ago

Discussion

ZeroWw

2 days ago

Great release! Please make also small ones. like 27B/A3B or 9B or something like that

Warsin

2 days ago

yea I can't even download this😭

nenor

2 days ago

a 8B or 35B-A5B would be nice

Rikunarita

2 days ago

I would love to see the release of smaller-backbone models (around 200B parameters or less) that incorporate mechanisms such as Engram or N-gram-based architectures.

Kimi-K3.1 Pro delivers benchmark results that surpass those of many frontier models. However, if a model cannot realistically be run locally, its value as an open-weight model becomes significantly limited for the average user.

Making such powerful models accessible in smaller, locally deployable configurations would be a huge step forward for the open-weight ecosystem.

binder11

2 days ago

Please make a small 1T model like Kimi K2 !!!!!!!!!!!!!!! that is a really nice size.

pathosethoslogos

2 days ago

Great comments! Please make me a convertible car, something like a Skyliner or something like that. A skyscraper would be nice too, like the Burj Khalifa.

Go make your own entitled models, buy your own compute, your own research team, data team, etc. etc. 🙄

Just be thankful that they're releasing anything.

localllmrig

1 day ago

Agree that local use is the real limit at this size. For anyone after something in the 27B to 35B range that runs today, this list shows which models fit one 24 GB card, with an estimated total per quant: https://localllmrig.com/vram/24gb?utm_source=hf&utm_campaign=fitlist-comment-2026-11-06

It counts 30 of 50 models at 4-bit. Qwen3.8-27B needs about 17.80 GB at UD-Q4_K_M and Qwen3.6-35B-A3B about 23.33 GB at UD-Q4_K_XL, both estimates. There is also a free email alert for when a new model beats the top picks.

Arkovski

about 12 hours ago

I want Kimi 3.1 Flash - 200-500B params with offloaded directory file to SSD.

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