As of: 2026-10-14 00:22 CEST.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/minicpm-v-4-7-35b-a3b/

# MiniCPM-V 4.7 35B-A3B

OpenBMB, MiniCPM-V family. Status: current. Type: multimodal model. Released: 2026-10-07. Training cutoff: not published.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Hugging Face | `openbmb/MiniCPM-V-4.7-35B-A3B` |  | https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B |
| Hugging Face (GGUF) | `openbmb/MiniCPM-V-4.7-35B-A3B-gguf` |  | https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B-gguf |
| Ollama | `openbmb/minicpm-v4.7` |  | https://ollama.com/openbmb/minicpm-v4.7 |

## Pricing

- Price not published.
- Last checked: 2026-10-13

## Specs

- Input: text, image, video
- Output: text
- Context window: not published
- Max output: not published
- Open weights: yes
- Licence: apache-2.0

## What it has not seen

- No published training cutoff. Counted from its release (2026-10-07): at least 166 AI events since (21 major, 2 historic), as of 2026-10-14.
- Everything since September 2026: https://postcutoff.com/since/2026-09/

## What stands out

- 35B mixture-of-experts model in phone memory: Vendor claim: 13+ tokens/s in 3 GB of memory on an iPhone 17 Pro Max, by loading only the experts about to be used (Lookahead Expert Scheduling), layer-wise quantization and DSpark speculative sampling. Not independently tested. Source: https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B

## Notes

35B total, about 3B active parameters; Qwen3.5/3.6 language backbone and SigLIP2 vision encoder. Image, multi-image and video input. A 1B version (openbmb/MiniCPM-V-4.7-1B-gguf and quantized builds) exists. Context length and benchmark numbers are not given in text on the model card.

Open vision-language model for on-device use.

```bash
vllm serve "openbmb/MiniCPM-V-4.7-35B-A3B"
# or: pip install "transformers[torch]>=5.18.0" torchvision torchcodec  (AutoModelForImageTextToText)
```

Model page: https://postcutoff.com/m/minicpm-v-4-7-35b-a3b/