Instructions to use tiny-random/kimi-k3.1-pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiny-random/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="tiny-random/kimi-k3.1-pro", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("tiny-random/kimi-k3.1-pro", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tiny-random/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 "tiny-random/kimi-k3.1-pro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/kimi-k3.1-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tiny-random/kimi-k3.1-pro

SGLang

How to use tiny-random/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 "tiny-random/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/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/kimi-k3.1-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "tiny-random/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/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/kimi-k3.1-pro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tiny-random/kimi-k3.1-pro with Docker Model Runner:
```
docker model run hf.co/tiny-random/kimi-k3.1-pro
```

Pack MXFP4 nibbles along the row, not across groups

by akopytko - opened about 10 hours ago

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

-0

Files changed (1) hide show

.meta/create_model.py +1 -0

.meta/create_model.py CHANGED Viewed

@@ -307,6 +307,7 @@ def create_model(source_model_id: str, source_revision: str, save_folder: str) -
             normalized = (blocks / scales.unsqueeze(-1)).clamp(-6, 6)
             codes = torch.bucketize(normalized.abs(), fp4_boundaries)
             codes += normalized.signbit() * 8
             state_dict[name.removesuffix(".weight") + ".weight_packed"] = (codes[:, ::2] | (codes[:, 1::2] << 4)).to(
                 torch.uint8
             )

             normalized = (blocks / scales.unsqueeze(-1)).clamp(-6, 6)
             codes = torch.bucketize(normalized.abs(), fp4_boundaries)
             codes += normalized.signbit() * 8
+            codes = codes.view(out_features, in_features)
             state_dict[name.removesuffix(".weight") + ".weight_packed"] = (codes[:, ::2] | (codes[:, 1::2] << 4)).to(
                 torch.uint8
             )