[2025.06.06] MiniCPM4 series are released! This model achieves ultimate efficiency improvements while maintaining optimal performance at the same scale! It can achieve over 5x generation acceleration on typical end-side chips! You can find technical report here.🔥🔥🔥
MiniCPM4 series are highly efficient large language models (LLMs) designed explicitly for end-side devices, which achieves this efficiency through systematic innovation in four key dimensions: model architecture, training data, training algorithms, and inference systems.
MiniCPM4-8B: The flagship of MiniCPM4, with 8B parameters, trained on 8T tokens. (<-- you are here)
MiniCPM4-0.5B: The small version of MiniCPM4, with 0.5B parameters, trained on 1T tokens.
MiniCPM4-8B-Eagle-FRSpec: Eagle head for FRSpec, accelerating speculative inference for MiniCPM4-8B.
MiniCPM4-8B-Eagle-FRSpec-QAT-cpmcu: Eagle head trained with QAT for FRSpec, efficiently integrate speculation and quantization to achieve ultra acceleration for MiniCPM4-8B.
MiniCPM4-8B-Eagle-vLLM: Eagle head in vLLM format, accelerating speculative inference for MiniCPM4-8B.
MiniCPM4-8B-marlin-Eagle-vLLM: Quantized Eagle head for vLLM format, accelerating speculative inference for MiniCPM4-8B.
BitCPM4-0.5B: Extreme ternary quantization applied to MiniCPM4-0.5B compresses model parameters into ternary values, achieving a 90% reduction in bit width.
BitCPM4-1B: Extreme ternary quantization applied to MiniCPM3-1B compresses model parameters into ternary values, achieving a 90% reduction in bit width.
MiniCPM4-Survey: Based on MiniCPM4-8B, accepts users' quiries as input and autonomously generate trustworthy, long-form survey papers.
MiniCPM4-MCP: Based on MiniCPM4-8B, accepts users' queries and available MCP tools as input and autonomously calls relevant MCP tools to satisfy users' requirements.
MiniCPM 4 is an extremely efficient edge-side large model that has undergone efficient optimization across four dimensions: model architecture, learning algorithms, training data, and inference systems, achieving ultimate efficiency improvements.
🏗️ Efficient Model Architecture:
InfLLM v2 -- Trainable Sparse Attention Mechanism: Adopts a trainable sparse attention mechanism architecture where each token only needs to compute relevance with less than 5% of tokens in 128K long text processing, significantly reducing computational overhead for long texts
🧠 Efficient Learning Algorithms:
Model Wind Tunnel 2.0 -- Efficient Predictable Scaling: Introduces scaling prediction methods for performance of downstream tasks, enabling more precise model training configuration search
BitCPM -- Ultimate Ternary Quantization: Compresses model parameter bit-width to 3 values, achieving 90% extreme model bit-width reduction
Efficient Training Engineering Optimization: Adopts FP8 low-precision computing technology combined with Multi-token Prediction training strategy
📚 High-Quality Training Data:
UltraClean -- High-quality Pre-training Data Filtering and Generation: Builds iterative data cleaning strategies based on efficient data verification, open-sourcing high-quality Chinese and English pre-training dataset UltraFinweb
UltraChat v2 -- High-quality Supervised Fine-tuning Data Generation: Constructs large-scale high-quality supervised fine-tuning datasets covering multiple dimensions including knowledge-intensive data, reasoning-intensive data, instruction-following data, long text understanding data, and tool calling data
⚡ Efficient Inference System:
CPM.cu -- Lightweight and Efficient CUDA Inference Framework: Integrates sparse attention, model quantization, and speculative sampling to achieve efficient prefilling and decoding
We recommend using CPM.cu for the inference of MiniCPM4. CPM.cu is a CUDA inference framework developed by OpenBMB, which integrates efficient sparse, speculative sampling, and quantization techniques, fully leveraging the efficiency advantages of MiniCPM4.
You can install CPM.cu by running the following command:
git clone https://github.com/OpenBMB/cpm.cu.git --recursive
cd cpm.cu
python3 setup.py install
MiniCPM4 natively supports context lengths of up to 32,768 tokens. To reproduce the long-text acceleration effect in the paper, we recommend using the LongRoPE factors that have been validated. Change the rope_scaling field in the config.json file as the following to enable LongRoPE.
After modification, you can run the following command to reproduce the long-context acceleration effect (the script will automatically download the model weights from HuggingFace)
python3 tests/test_generate.py
For more details about CPM.cu, please refer to the repo CPM.cu.
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
torch.manual_seed(0)
path = 'openbmb/MiniCPM4-8B'
device = "cuda"
tokenizer = AutoTokenizer.from_pretrained(path)
model = AutoModelForCausalLM.from_pretrained(path, torch_dtype=torch.bfloat16, device_map=device, trust_remote_code=True)
# User can directly use the chat interface
# responds, history = model.chat(tokenizer, "Write an article about Artificial Intelligence.", temperature=0.7, top_p=0.7)
# print(responds)
# User can also use the generate interface
messages = [
{"role": "user", "content": "Write an article about Artificial Intelligence."},
]
prompt_text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
)
model_inputs = tokenizer([prompt_text], return_tensors="pt").to(device)
model_outputs = model.generate(
**model_inputs,
max_new_tokens=1024,
top_p=0.7,
temperature=0.7
)
output_token_ids = [
model_outputs[i][len(model_inputs[i]):] for i in range(len(model_inputs['input_ids']))
]
responses = tokenizer.batch_decode(output_token_ids, skip_special_tokens=True)[0]
print(responses)
MiniCPM4-8B supports InfLLM v2, a sparse attention mechanism designed for efficient long-sequence inference. It requires the infllmv2_cuda_impl library.
You can install it by running the following command:
These parameters control the behavior of InfLLM v2:
kernel_size (default: 32): The size of semantic kernels.
kernel_stride (default: 16): The stride between adjacent kernels.
init_blocks (default: 1): The number of initial blocks that every query token attends to. This ensures attention to the beginning of the sequence.
block_size (default: 64): The block size for key-value blocks.
window_size (default: 2048): The size of the local sliding window.
topk (default: 64): The specifies that each token computes attention with only the top-k most relevant key-value blocks.
use_nope (default: false): Whether to use the NOPE technique in block selection for improved performance.
dense_len (default: 8192): Since Sparse Attention offers limited benefits for short sequences, the model can use standard (dense) attention for shorter texts. The model will use dense attention for sequences with a token length below dense_len and switch to sparse attention for sequences exceeding this length. Set this to -1 to always use sparse attention regardless of sequence length.
MiniCPM4 natively supports context lengths of up to 32,768 tokens. For conversations where the total length (including both input and output) significantly exceeds this limit, we recommend using RoPE scaling techniques for effective handling of long texts. We have validated the model's performance on context lengths of up to 131,072 tokens by modifying the LongRoPE factor.
You can apply the LongRoPE factor modification by modifying the model files. Specifically, in the config.json file, adjust the rope_scaling fields.
Also, you can start the inference server by running the following command:
Note: In vLLM's chat API, add_special_tokens is False by default. This means important special tokens—such as the beginning-of-sequence (BOS) token—will not be added automatically. To ensure the input prompt is correctly formatted for the model, you should explicitly set extra_body={"add_special_tokens": True}.
vllm serve openbmb/MiniCPM4-8B
Then you can use the chat interface by running the following code:
import openai
client = openai.Client(base_url="http://localhost:8000/v1", api_key="EMPTY")
response = client.chat.completions.create(
model="openbmb/MiniCPM4-8B",
messages=[
{"role": "user", "content": "Write an article about Artificial Intelligence."},
],
temperature=0.7,
max_tokens=1024,
extra_body=dict(add_special_tokens=True), # Ensures special tokens are added for chat template
)
print(response.choices[0].message.content)
On two typical end-side chips, Jetson AGX Orin and RTX 4090, MiniCPM4 demonstrates significantly faster processing speed compared to similar-size models in long text processing tasks. As text length increases, MiniCPM4's efficiency advantage becomes more pronounced. On the Jetson AGX Orin platform, compared to Qwen3-8B, MiniCPM4 achieves approximately 7x decoding speed improvement.
MiniCPM4 is pre-trained on 32K long texts and achieves length extension through YaRN technology. In the 128K long text needle-in-a-haystack task, MiniCPM4 demonstrates outstanding performance.
import os
os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com"
from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("openbmb/MiniCPM4-8B", revision="bb2ae14cf59d4ca769c4e42ece54cc3b82a58ef7")
model = AutoModelForCausalLM.from_pretrained("openbmb/MiniCPM4-8B", revision="bb2ae14cf59d4ca769c4e42ece54cc3b82a58ef7", device_map="auto")
