MiniCPM-V 4.7 1B on Other Android device (4 GB RAM)

Can the Other Android device (4 GB RAM) run MiniCPM-V 4.7 1B?

Yes, it runs well

Runs well. At Q3_K_M it needs about 1.5 GB of 1.8 GB usable, with an estimated speed of 25.7 tok/s.

Best quant: Q3_K_M
Generation speed: 25.7 tok/s
Prompt processing: 44.5 tok/s
First token (1K prompt): 23.0 s
Largest safe context: 35K

Best app to run it on the Other Android device (4 GB RAM)

On Android

PocketPal AI

Mobile app · runs on llama.cpp

Runs the model on the device itself, offline: download it once in the app.

Get it on Google Play Then download MiniCPM-V 4.7 1B in the app

ChatterUI Runs the model on the device itself, offline: download it once in the app.

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download MiniCPM-V 4.7 1B

On a phone, tablet or other mobile device you install a chat app and get the model file through it, so there are no commands to copy. Use MiniCPM-V-4_7-1B-IQ3_M.gguf for the Other Android device (4 GB RAM).

On Android

PocketPal AI

Free mobile app · runs the model on the device with llama.cpp

Get it on Google Play

Install PocketPal AI (free) and open it.
Tap the menu (☰), then Models, then + and Add from Hugging Face.
Search for openbmb/MiniCPM-V-4.7-1B-gguf, open it and download MiniCPM-V-4_7-1B-IQ3_M.gguf.
Tap Load next to the model and start a chat. It runs on the device, with no connection needed from then on.

Also works: ChatterUI

Files and commands for a computer

Recommended for the Other Android device (4 GB RAM): Q3_K_M. Every other quantization has its own download link in the table below.

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q3_K_M (IQ3_M) file (opens in a new tab)

huggingface.co/openbmb/MiniCPM-V-4.7-1B-gguf (opens in a new tab)

ollama run hf.co/openbmb/MiniCPM-V-4.7-1B-gguf:IQ3_M

Original weights · Hugging Face

Full-precision files from OpenBMB, for fine-tuning, vLLM, Transformers.

huggingface.co/openbmb/MiniCPM-V-4.7-1B (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from OpenBMB. The weights themselves download from Hugging Face.

github.com/OpenBMB/MiniCPM-V (opens in a new tab)

git clone https://github.com/OpenBMB/MiniCPM-V

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q3_K_M, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs well: How well it runs on the Other Android device (4 GB RAM): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	15.1 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	12.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Match score	67.1 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the Other Android device (4 GB RAM)

Quant	Memory needed	Available	Speed	Runs	Quality	File to pick in the app
`Q2_K`	1.3 GB	1.8 GB	32.6 tok/s	Runs great	28Basic quality (D) · 28 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	—
`Q3_K_M`	1.5 GB	1.8 GB	25.7 tok/s	Runs well	33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking.	`IQ3_M`
`Q4_K_M`	1.6 GB	1.8 GB	19.9 tok/s	Runs welltight fit	38Basic quality (D) · 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	`Q4_K_M`
`Q5_K_M`	1.8 GB	1.8 GB	—	Won't runover the app memory limit	38Basic quality (D) · 38 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking.	`Q5_K_M`
`Q6_K`	2.0 GB	1.8 GB	—	Won't runover the app memory limit	39Basic quality (D) · 39 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking.	`Q6_K`
`Q8_0`	2.3 GB	1.8 GB	—	Won't runover the app memory limit	40Fair quality (C) · 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking.	`Q8_0`
`F16`	3.8 GB	1.8 GB	—	Won't runover the app memory limit	40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 5/100 (quality, fit, speed and memory use) sets the ranking.	`F16`

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4060 F16 · 50.5 tok/s · Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. Intel Arc B580 F16 · 64.7 tok/s · Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3060 12GB F16 · 66.9 tok/s · Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4070 SUPER F16 · 93.6 tok/s · Runs great
40Fair quality (C) · 40 of 100: how capable the model is at F16, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Pro (24 GB) F16 · 43.7 tok/s · Runs great
40Fair quality (C) · 40 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 (16 GB) Q8_0 · 41.7 tok/s · Runs great

Quick answers

How much memory does MiniCPM-V 4.7 1B need on the Other Android device 4GB?

At Q3_K_M it needs about 1.5 GB. The Other Android device 4GB has 1.8 GB usable.

How fast is MiniCPM-V 4.7 1B on the Other Android device 4GB?

About 25.7 tok/s at Q3_K_M, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of MiniCPM-V 4.7 1B should I use on the Other Android device 4GB?

Q3_K_M. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.