Skip to content
Find local AI

Your computer: not set · Set it

Photos & Images

MiniCPM-V 4.7 1B

Apache 2.0

OpenBMB's small vision model for pictures, documents, and video. It is not a mixture of experts.

PhotosPhoto

Compare two models

looks at pictures, documents, and video in a small file

The Q4_K_M language file is about 0.6 GB, and the vision projector brings the running size to about 1.7 GB. The 3 GB phone figure belongs to the 35B-A3B size.

0.6 GB · · needs ~1.7 GB · smallest file

Runs in:

Fits on

Laptops

Desktops

Macs

Mini PCs & servers

Small devices

Flavors:File · 0.6 GB

Tasks this model can do

How good is it?

OpenBMB's small vision model for pictures, documents, and video. It is not a mixture of experts.

This job is not on the July 2026 chat board, so the line above comes from the model card.

Clean printed pages come out well. Handwriting, messy photos, and tiny type are weaker. Tesseract and PaddleOCR are light. olmOCR, Qwen2.5-VL, and MiniCPM-V 4.7 read the page more like a person. The larger MiniCPM-V file needs about 21 GB while it runs.

Will it fit?

A green light means it fits and feels fine. Amber means it fits, with a wait or a tight memory pool. Red means the memory it needs is bigger than the room we count.

HONEST COMPARISON

Local or cloud: Describe a video?

Reviewed Oct 2026

  • Privacy: Local winsThe clip stays on your computer
  • Cost: TieFree options exist on both sides
  • Quality: Cloud winsFewer missed moments on long clips
  • Ease: Cloud winsDrop the file in a chat

Running MiniCPM-V 4.7 1B locally

A local vision model can describe a clip you already have, without uploading it.

  • The clip stays on disk
  • Works offline
  • Can mention on-screen text and scene changes

Where it falls short

  • A long clip is sampled, so it can miss a short moment
  • It can invent a detail that is not in the frames
See local tools for this task ↓

Cloud & apps

Cloud video tools are easier to open and usually stronger on long or cluttered clips.

  • CChatGPTCloud AIFreemium

    A short clip can come back as a description in the same chat.

    Best for: Short clipsFollow-up questions

    The video is uploaded

    Visit ChatGPT (opens in new tab)
  • CClaudeCloud AIFreemium

    Careful wording when the description should stay literal.

    Best for: Literal descriptions

    The video is uploaded

    Visit Claude (opens in new tab)
  • GGeminiCloud AIFreemiumPopular choice

    Describes a video you upload, including longer clips.

    Best for: Longer clipsQuick looks

    The video is sent to Google

    Visit Gemini (opens in new tab)

Our take

Describe a private clip locally. Use a cloud video model when the clip is long and it can be shared.

Choose local if…

  • The video is private
  • You are offline

Choose cloud if…

  • The clip is long
  • You want the least setup

Will it run on my computer?

Pick the memory you have. A Mac counts 75% of unified memory. A graphics card counts 90% of its memory. A PC without a graphics card counts half of its RAM. The memory we plan around is 1.7 GB.

MiniCPM-V 4.7 1B fits. About 12 GB is free, and this model needs 1.7 GB.

Official install guides

We link to the makers. We do not write our own install steps.