Instructions to use AXERA-TECH/MiniCPM-V-4.6-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AXERA-TECH/MiniCPM-V-4.6-GPTQ with Transformers:

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

pipe = pipeline("image-text-to-text", model="AXERA-TECH/MiniCPM-V-4.6-GPTQ", device_map="auto")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("AXERA-TECH/MiniCPM-V-4.6-GPTQ", dtype="auto", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AXERA-TECH/MiniCPM-V-4.6-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AXERA-TECH/MiniCPM-V-4.6-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AXERA-TECH/MiniCPM-V-4.6-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AXERA-TECH/MiniCPM-V-4.6-GPTQ

SGLang

How to use AXERA-TECH/MiniCPM-V-4.6-GPTQ 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 "AXERA-TECH/MiniCPM-V-4.6-GPTQ" \
    --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": "AXERA-TECH/MiniCPM-V-4.6-GPTQ",
		"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 "AXERA-TECH/MiniCPM-V-4.6-GPTQ" \
        --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": "AXERA-TECH/MiniCPM-V-4.6-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AXERA-TECH/MiniCPM-V-4.6-GPTQ with Docker Model Runner:
```
docker model run hf.co/AXERA-TECH/MiniCPM-V-4.6-GPTQ
```

yongqiang commited on Jun 9

Commit

3fc0acb

1 Parent(s): 9074c52

Update README performance metrics

Browse files

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -37,15 +37,15 @@ This package has been validated on the following AX650-based device:
 ## Performance
-All measurements below were taken on AX650 / NPU3. `TTFT` stands for time to first token.
-The text-only smoke prompt was kept within one `128`-token prefill chunk. The image row was measured with the packaged fixed-shape `448x448` vision encoder and `assets/sample.png`. The video row used the packaged sample video with `video:assets/red-panda-openai.mp4:2`.
 | Scenario | Input tokens | Prefill chunks | TTFT | Decode |
 |---|---:|---:|---:|---:|
-| Text-only smoke prompt | `25` | `1 x 128` | `223.75 ms avg` (`223.05-225.72 ms`) | `n/a (single-token response)` |
-| Image prompt | `88` | `1 x 128` | `218.40 ms avg` (`217.58-219.26 ms`) | `24.55 token/s avg` |
-| Video prompt | `1272` | `10 x 128` | `2326.39 ms avg` (`2314.59-2338.18 ms`) | `22.71 token/s avg` |
 The packaged runtime uses the following context layout:
@@ -53,7 +53,7 @@ The packaged runtime uses the following context layout:
 - `kv_cache_len=2047`
 - `prefill_max_token_num=1280`
-`Input tokens` in the table above refers to the full request length after chat templating, not just the visual soft tokens. For the shipped `448x448` vision encoder, each selected image block contributes `64` visual soft tokens. Under the current packaged runtime settings, the sample video request in this README uses `1272` total input tokens and spans `10` prefill chunks.
 ### Startup Runtime Footprint

 ## Performance
+All measurements below were taken on AX650 / NPU3. `TTFT` stands for time to first token. In this table, `TTFT` is measured end-to-end from request arrival at `axllm serve` to the first generated token, so the multimodal rows include media preprocessing and vision encoding time.
+The text-only smoke prompt was kept within one `128`-token prefill chunk. To avoid one-time startup effects, the text row below excludes the first request after service startup. Its `Decode` figure was measured with longer text-only generations (`max_tokens=256`) to better reflect sustained decode throughput; the short smoke reply used for the `TTFT` row is effectively a single-token answer and would otherwise under-report decode speed. The image row was measured with the packaged fixed-shape `448x448` vision encoder and `assets/sample.png`. The video row used the packaged sample video with `video:assets/red-panda-openai.mp4:2`.
 | Scenario | Input tokens | Prefill chunks | TTFT | Decode |
 |---|---:|---:|---:|---:|
+| Text-only smoke prompt | `25` | `1 x 128` | `260.81 ms avg` (`259.01-262.61 ms`) | `24.07 token/s avg` |
+| Image prompt | `88` | `1 x 128` | `719.79 ms avg` (`708.57-732.47 ms`) | `24.49 token/s avg` |
+| Video prompt | `1271` | `10 x 128` | `9555.33 ms avg` (`9484.00-9647.62 ms`) | `23.87 token/s avg` |
 The packaged runtime uses the following context layout:
 - `kv_cache_len=2047`
 - `prefill_max_token_num=1280`
+`Input tokens` in the table above refers to the full request length after chat templating, not just the visual soft tokens. For the shipped `448x448` vision encoder, each selected image block contributes `64` visual soft tokens. Under the current packaged runtime settings, the sample video request in this README uses `1271` total input tokens and spans `10` prefill chunks.
 ### Startup Runtime Footprint