MiniCPM5-1B

MiniCPM Text Series

MiniCPM5-1B is the first model in the MiniCPM5 series, built for local assistants, coding agents, tool-calling workflows and inference scenarios that need a compact model. It offers native long-context capability at a low deployment cost and supports both Think and No Think chat modes from the same weights.

Choose the BF16, GGUF or MLX weights that suit your runtime environment; for training or research, you can also look at the SFT and Base versions.

Model Downloads

Currently recommended version: MiniCPM5-1B (BF16 / GGUF / MLX):

HuggingFace	ModelScope
MiniCPM5-1B	MiniCPM5-1B
MiniCPM5-1B-SFT	MiniCPM5-1B-SFT
MiniCPM5-1B-Base	MiniCPM5-1B-Base
MiniCPM5-1B-GGUF	MiniCPM5-1B-GGUF
MiniCPM5-1B-MLX	MiniCPM5-1B-MLX

Quick Start

For the three most common inference paths, you can start a service or run local inference as shown below.

vLLM (OpenAI-compatible service, NVIDIA GPU)

Bash

pip install "vllm>=0.21" && vllm serve openbmb/MiniCPM5-1B --port 8000

Test request:

Bash

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openbmb/MiniCPM5-1B",
    "messages": [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}],
    "max_tokens": 128,
    "temperature": 0.7
  }'

SGLang (OpenAI-compatible service, NVIDIA GPU)

Bash

pip install "sglang[srt]>=0.5.12" && python -m sglang.launch_server --model-path openbmb/MiniCPM5-1B --port 30000

Test request:

Bash

curl http://localhost:30000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openbmb/MiniCPM5-1B",
    "messages": [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}],
    "max_tokens": 128,
    "temperature": 0.7
  }'

Transformers (local Python inference, CPU or GPU)

Bash

pip install -U "transformers>=5.6" accelerate torch

Local inference test:

Python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "openbmb/MiniCPM5-1B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}]
inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    enable_thinking=False,
    return_tensors="pt",
).to(model.device)

outputs = model.generate(inputs, max_new_tokens=128)
print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))

Recommended chat template sampling parameters:

Mode	Recommended sampling parameters	How to enable
Think	temperature=0.9, top_p=0.95	enable_thinking=True
No Think	temperature=0.7, top_p=0.95	enable_thinking=False

For the other backends (llama.cpp / Ollama / LM Studio / MLX), see the cookbook table.

Tool Calling (Function Call)

SGLang is recommended for tool calling: MiniCPM5-1B emits tool calls in XML format, and SGLang's built-in minicpm5 parser automatically converts them into the OpenAI-compatible tool_calls field.

Bash

python -m sglang.launch_server --model-path openbmb/MiniCPM5-1B --port 30000 \
    --tool-call-parser minicpm5      # or: --tool-call-parser auto

Related Resources

Topic	Link
Deployment	MiniCPM5-1B deployment guide
Fine-tuning	MiniCPM5-1B fine-tuning guide
Code repository	OpenBMB/MiniCPM
Agent Skills	Deployment and fine-tuning skills