Instructions to use moonshotai/Kimi-Linear-48B-A3B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moonshotai/Kimi-Linear-48B-A3B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="moonshotai/Kimi-Linear-48B-A3B-Instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("moonshotai/Kimi-Linear-48B-A3B-Instruct", trust_remote_code=True, dtype="auto")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use moonshotai/Kimi-Linear-48B-A3B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moonshotai/Kimi-Linear-48B-A3B-Instruct"
# 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-Linear-48B-A3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/moonshotai/Kimi-Linear-48B-A3B-Instruct

SGLang

How to use moonshotai/Kimi-Linear-48B-A3B-Instruct 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-Linear-48B-A3B-Instruct" \
    --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-Linear-48B-A3B-Instruct",
		"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 "moonshotai/Kimi-Linear-48B-A3B-Instruct" \
        --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-Linear-48B-A3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use moonshotai/Kimi-Linear-48B-A3B-Instruct with Docker Model Runner:
```
docker model run hf.co/moonshotai/Kimi-Linear-48B-A3B-Instruct
```

No space to test it, not on open router, are there obstacles ?

by cmp-nct - opened Nov 2, 2025

Discussion

cmp-nct

Nov 2, 2025

Just wondering if the model is possibly having issues or performing horrible ?
I wanted to test it but it's not supported on llama.cpp, is not on openrouter and doesn't have a single space on HF.

fimbulvntr

Nov 2, 2025

It's a bit too recent, released mere days ago
More of an architecture demonstrator than a model meant for production use (though it might make a great base for FT)
Requires VLLM nightly, and the VLLM implementation was broken on release day (and AFAIK still broken to this day)

I think it should be a little better once VLLM releases next (should be any day now!)

zhiyuan8

Nov 3, 2025

There is a related fix PR: https://github.com/vllm-project/vllm/pull/27905

Documentation here: https://docs.vllm.ai/projects/recipes/en/latest/moonshotai/Kimi-Linear.html#4-gpu-tensor-parallel
And I have tested latest commit: https://github.com/vllm-project/vllm/commit/1bf43ae35d7f6a83cc2025b8c0a2332456f4afe9

|      Tasks      |Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k_opencompass|      3|flexible-extract|     0|exact_match|↑  |0.9075|±  | 0.008|

zhiyuan8 changed discussion status to closed Nov 3, 2025

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