Parameters
1B
Context Length
262K
Modality
Text
Architecture
Dense
KV Cache Rate
48 KB/tok
License
Apache 2.0
Release Date
6 Oct 2026
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
Precision
Context
Workload Memory: 3.34 GB
INT4 · 8K context
Base Weights: 0.47 GB
KV Cache: 0.38 GB
Runtime & Graphs: 2.50 GB
1x NVIDIA L4 24GB
3.3 / 21.6 GB usable
1x RTX 3060 12GB
3.3 / 10.8 GB usable
Apple M4 24GB
3.3 / 18.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Checkpoint format
FP16
CUDA
>= 11.8
PyTorch
>= 2.1
Transformers
>= 5.18.0
| Architecture | Precision | Attention kernel |
|---|---|---|
Blackwell, Hopper | Native FP16 | FlashAttention-3 |
Ada Lovelace, Ampere | Native FP16 | FlashAttention-2 |
No evaluation benchmarks for MiniCPM-V 4.7 1B available.
Overall Rank
-
Coding Rank
-
MiniCPM-V 4.7 1B is an ultra-compact multimodal model tailored for edge deployment and on-device vision-language tasks. It delivers competitive optical character recognition and multimodal conversational performance despite its small footprint.
Attention
Attention Structure
Grouped-Query Attention
Attention Heads
8
Key-Value Heads
2
Attention Head Dimension
256
Position Embedding
ROPE
RoPE Theta
10,000,000
Linear Attention
Yes
Linear Attention Ratio
75.0%
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Base Precision
FP16
Hidden Dimension Size
1,024
Number of Layers
24
FFN Intermediate Size (Dense)
3,584
Multi-Token Prediction Heads
1
Tokenizer
Vocabulary Size
248,094
The MiniCPM model family developed by OpenBMB.