Instructions to use fwerkor/MiniCPM5-2B-Diffusion-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fwerkor/MiniCPM5-2B-Diffusion-Base with Transformers:

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

pipe = pipeline("text-generation", model="fwerkor/MiniCPM5-2B-Diffusion-Base", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fwerkor/MiniCPM5-2B-Diffusion-Base", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("fwerkor/MiniCPM5-2B-Diffusion-Base", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fwerkor/MiniCPM5-2B-Diffusion-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fwerkor/MiniCPM5-2B-Diffusion-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fwerkor/MiniCPM5-2B-Diffusion-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/fwerkor/MiniCPM5-2B-Diffusion-Base

SGLang

How to use fwerkor/MiniCPM5-2B-Diffusion-Base 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 "fwerkor/MiniCPM5-2B-Diffusion-Base" \
    --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": "fwerkor/MiniCPM5-2B-Diffusion-Base",
		"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 "fwerkor/MiniCPM5-2B-Diffusion-Base" \
        --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": "fwerkor/MiniCPM5-2B-Diffusion-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fwerkor/MiniCPM5-2B-Diffusion-Base with Docker Model Runner:
```
docker model run hf.co/fwerkor/MiniCPM5-2B-Diffusion-Base
```

MiniCPM5-2B-Diffusion-Base / diffusion_config.json

fwerkor

Release MiniCPM5-2B-Diffusion-Base

5f58db9 verified 4 days ago

Raw

History Blame Contribute Delete

1.74 kB

	{
	"format": "cid-diffusion-base-v1",
	"base_model": "openbmb/MiniCPM5-2B-Base",
	"objective": "LLaDA-style masked diffusion",
	"attention": "bidirectional",
	"mask_token": "<\|cid_mask\|>",
	"mask_token_id": 130560,
	"sequence_length": 2048,
	"completed_steps": 5087,
	"global_batch_size": 96,
	"tokens_seen": 1000144896,
	"training_corpus": {
	"format": "cid-diffusion-stream-v1",
	"base_model": "openbmb/MiniCPM5-2B-Base",
	"sequence_length": 2048,
	"sources": [
	{
	"name": "ultrax-web-en",
	"repo": "openbmb/UltraX-Preview",
	"revision": "a88527587389fd4ab352e9ad1273f4c0a234d8df",
	"pattern": "data/UltraX-Ultra-FineWeb/*.parquet",
	"column": "cleaned_content",
	"weight": 0.7
	},
	{
	"name": "ultrafineweb-zh",
	"repo": "openbmb/Ultra-FineWeb",
	"revision": "02c85641e3d19a854be2e09139c25adaa9518063",
	"pattern": "data/ultrafineweb_zh/*.parquet",
	"column": "content",
	"weight": 0.15
	},
	{
	"name": "ultradata-code",
	"repo": "openbmb/UltraData-Code",
	"revision": "85182d829f2ce7ea07cca72ebfc509deea1d9f5f",
	"pattern": "data/UltraData-Code-L2//.parquet",
	"column": "content",
	"weight": 0.1
	},
	{
	"name": "ultradata-math",
	"repo": "openbmb/UltraData-Math",
	"revision": "fe10db8efd35597fd7fcff8ff576b5ec4ea5ff87",
	"pattern": "data/UltraData-Math-L2-preview/*.parquet",
	"column": "content",
	"weight": 0.05
	}
	]
	},
	"mask_ratio_range": [
	0.001,
	1.0
	],
	"hf_loader": "CIDDiffusionForMaskedLM",
	"hf_auto_model": "AutoModelForCausalLM",
	"requires_trust_remote_code": true
	}