Model Detail

MiniCPM-V-4.6 is a multimodal model with 650M parameters released by openbmb. The model is registered under the image-text-to-text pipeline tag on Hugging Face, distributed under the permissive apache-2.0 license.

MiniCPM-V-4.6 ships with 650M parameters. Total weight footprint is approximately 1.3 GB, which is the relevant figure when planning local-inference VRAM. The apache-2.0 license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.

Downloads of MiniCPM-V-4.6 have moved +33.1% over the trailing seven days. That puts the model in active uptrend territory; a sustained move of this size usually reflects a recent release, a viral integration, or a benchmark surprise rather than steady-state demand. These numbers are signal, not guarantee — week-over-week download counts on Hugging Face also reflect mirror traffic, CI scrapes, and one-off benchmarking runs.

MiniCPM-V-4.6 is best fit for mixed text-and-image reasoning tasks such as document understanding. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.