# 35B Vision Model Runs on iPhone at 13 Tokens a Second, OpenBMB Says

Published 2026-10-13 · AI.info Team · AI.info

> OpenBMB released MiniCPM-V 4.7 on October 12, including a 35B-A3B mixture-of-experts vision model and a smaller 1B version. Speeds above 13 tokens per second on an iPhone 17 Pro Max are reported by the company, while its phone-app guide calls for 12 GB of device RAM and about 17 GB of storage.

More than 13 tokens per second on an iPhone is [OpenBMB](https://github.com/OpenBMB/MiniCPM-V)’s headline claim for MiniCPM-V 4.7, a vision-language model with 35 billion total parameters. Its 35B-A3B mixture-of-experts version reached that decoding speed on an iPhone 17 Pro Max while using 3 GB of memory, the company says. OpenBMB released the model on October 12, alongside a separate 1B version.

## How OpenBMB gets a 35B model onto a phone

MiniCPM-V 4.7 handles images, multiple images, documents and video, according to its release documentation. OpenBMB describes the 35B-A3B version as the first MoE model in the MiniCPM-V family; the “A3B” designation refers to 3 billion active parameters, rather than all 35 billion being used for each token. The model also supports mixed 4× and 16× visual-token compression, which lets developers trade image detail for lower computation.

The speed claim rests on an inference setup designed around the constraints of phone storage and memory. OpenBMB says the model is “achieving a decoding speed of 13+ tokens/s with just 3 GB of memory on iPhone 17 Pro Max.” The company credits Lookahead Expert Scheduling, which predicts and preloads the experts the next layer is likely to need, along with weight quantization and an optimized storage layout that shrink the footprint and speed up loading.

## The 3 GB figure is not the full phone requirement

OpenBMB’s companion app repository adds an important qualification: the phone deployment streams expert weights from an SSD instead of keeping them all in memory. Its guide lists 12 GB of device RAM as the recommendation for the 35B-A3B version and says the setup has been validated on a 12 GB iPhone 17 Pro Max. The guide estimates a 16.6 GB download and about 17 GB of free storage, including the main model file, vision projector and 40 expert shards.

That distinction matters: the 3 GB figure in the model announcement is not a claim that the full model download fits in 3 GB of phone storage, or that any 3 GB phone can run it. The phone app uses a split, on-device package and streams expert weights from storage; OpenBMB says phones with less RAM fall back to a smaller expert cache and run considerably more slowly. The special split package is made for the demo, rather than a general-purpose model release, the guide also says.

## OpenBMB pairs the MoE with a 1B model

The smaller MiniCPM-V 4.7 variant is a 1B dense model, not an MoE. The smaller model outperforms its earlier MiniCPM-V 4.6 and Qwen3.5 0.8B on the company’s reported performance and efficiency comparisons, while the 35B-A3B model scores higher and runs in less inference time than Qwen3.6 35B-A3B in Instruct mode. The release documentation does not provide numerical benchmark results in its text, so those comparisons remain the developers’ claims.

The release supports deployment on iOS, Android and HarmonyOS, and lists chip support spanning Arm, Intel, AMD, Rockchip and Houmo. The code and model weights are available under the Apache-2.0 license. Its companion app repository provides sample phone apps and the specialist storage arrangement needed to test the 35B model locally.

## Sources

1. Original source [OpenBMB GitHub · page 1](https://github.com/OpenBMB/MiniCPM-V) github.com

Other pages read

2. [OpenBMB GitHub · Minicpm V Apps](https://github.com/OpenBMB/MiniCPM-V-Apps) github.com

## Related stories

- [PrismML Fits a 27B Multimodal Model Into 5.9GB](https://ai.info/news/prismml-fits-27b-multimodal-model-into-5-9gb)
- [Google’s TranslateGemma Brings Offline Translation to 55 Languages](https://ai.info/news/google-translategemma-offline-translation-55-languages)
- [Liquid AI Releases LFM2.5, a 24B Model for Local Agents](https://ai.info/news/liquid-ai-lfm25-24b-local-agents)
- [Goodfire Uses Model Probes to Cut AI Agent Monitoring Costs](https://ai.info/news/goodfire-probes-cut-ai-agent-monitoring-costs)
- [Google Opens Gemini 4 Argon to Developers and Businesses](https://ai.info/news/google-gemini-4-argon-developers-enterprise)

## Explore

[More articles](https://ai.info/news)

---

Canonical: https://ai.info/news/openbmb-minicpm-v-47-35b-moe-iphone
