Instructions to use amd/Kimi-K2.5-W4A8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use amd/Kimi-K2.5-W4A8 with Transformers:

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

pipe = pipeline("image-text-to-text", model="amd/Kimi-K2.5-W4A8", 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("amd/Kimi-K2.5-W4A8", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use amd/Kimi-K2.5-W4A8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amd/Kimi-K2.5-W4A8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amd/Kimi-K2.5-W4A8",
		"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/amd/Kimi-K2.5-W4A8

SGLang

How to use amd/Kimi-K2.5-W4A8 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 "amd/Kimi-K2.5-W4A8" \
    --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": "amd/Kimi-K2.5-W4A8",
		"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 "amd/Kimi-K2.5-W4A8" \
        --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": "amd/Kimi-K2.5-W4A8",
		"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 amd/Kimi-K2.5-W4A8 with Docker Model Runner:
```
docker model run hf.co/amd/Kimi-K2.5-W4A8
```

Model Overview

Model Architecture: Kimi-K2.5
- Input: Text
- Output: Text
Supported Hardware Microarchitecture: AMD MI300/MI325/MI350/MI355
ROCm: 7.1.0
Operating System(s): Linux
Inference Engine: vLLM
Model Optimizer: AMD-Quark
- Weight quantization: MOE-only, INT4 Per-Channel & FP8E4M3, Static
- Activation quantization: MOE-only, FP8E4M3, Dynamic

This model was built with Kimi-K2.5 model by applying AMD-Quark for INT4-FP8 quantization.

Model Quantization

The model was quantized from moonshotai/Kimi-K2.5 using AMD-Quark. The weights and activations are quantized to INT4-FP8.

Deployment

Use with vLLM

This model can be deployed efficiently using the vLLM backend.

Evaluation

The model was evaluated on GSM8K benchmarks.

Accuracy

Benchmark	Kimi-K2.5	Kimi-K2.5-W4A8(this model)	Recovery
GSM8K (flexible-extract)	94.09	93.40	99.27%

Reproduction

The GSM8K results were obtained using the lm-evaluation-harness framework, based on the Docker image vllm/vllm-openai-rocm:v0.14.0.

Install the vLLM (commit ecb4f822091a64b5084b3a4aff326906487a363f) and lm-eval (Version: 0.4.10) in container first.

git clone https://github.com/vllm-project/vllm.git
cd vllm
python3 setup.py develop

pip install lm-eval

Launching server

VLLM_ROCM_USE_AITER_MLA=0 VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_FUSION_SHARED_EXPERTS=0 VLLM_ROCM_USE_AITER_FP4BMM=0 vllm serve amd/Kimi-K2.5-W4A8 \
  --tensor-parallel-size 8 \
  --mm-encoder-tp-mode data \
  --tool-call-parser kimi_k2 \
  --reasoning-parser kimi_k2 \
  --trust-remote-code \
  --enforce-eager

Evaluating model in a new terminal

lm_eval \
  --model local-completions \
  --model_args "model=amd/Kimi-K2.5-W4A8,base_url=http://0.0.0.0:8000/v1/completions,tokenized_requests=False,tokenizer_backend=None,num_concurrent=32" \
  --tasks gsm8k \
  --num_fewshot 5 \
  --batch_size 1

License

Downloads last month: 1,727

Safetensors

Model size

139B params

Tensor type

BF16

F32

I32

Model tree for amd/Kimi-K2.5-W4A8

Base model

moonshotai/Kimi-K2.5

Quantized

(38)

this model