How much VRAM does MiniCPM-V 4.7 35B-A3B need?

About 22.2 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 22.8 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.

Will MiniCPM-V 4.7 35B-A3B run on a 32GB GPU?

Yes — at Q4_K_M and 4K context it needs about 22.2 GB, which leaves 9.8 GB spare on a RTX 5090, or on any of the other 32 GB cards here (4 in all). That is the smallest size in this index that clears it comfortably; 26 of 78 do.

Which quantization of MiniCPM-V 4.7 35B-A3B should I use?

No published quality comparison exists for this model, so pick by footprint: Q4_K_M is the level most people run, and a higher bits-per-weight level is more faithful. All 4 levels in the index are Q4_K_S, Q4_K_M, Q5_K_M, Q8_0. Note this is a mixture-of-experts model — all parameters must be resident even though only a fraction are active per token, so the memory cost follows the total, not the active count.

MiniCPM-V 4.7 35B-A3B

35B-A3B

OpenBMB MiniCPM-V

A mixture-of-experts vision model from OpenBMB. It reads images and video on top of a 34.7B language model with about 3B parameters active per token, the same layout as Qwen3.6 35B-A3B. Only 10 of its 40 layers keep a growing KV cache, so context is cheap. At Q4_K_S it needs about 20.9 GB at 4K, comfortable on a 24 GB card; Q4_K_M (22.2 GB) is tight there, and a 32 GB card holds Q4_K_M across the whole 256K window. The vision encoder is a separate 1.1 GB file on top of these figures, which turns Q4_K_S on 24 GB tight as well once images are loaded. llama.cpp runs the text model as `qwen35moe`. Sizes come from OpenBMB's own GGUF files, and no per-level quality loss has been published.

⬇ 95.2K HF downloads♥ 1079 likesopenbmb/MiniCPM-V-4.7-35B-A3B· stats from 10/20/2026

Consumer GPUPro GPUMac / Apple Silicon

262K

Max Context

Quant Variants

GGUF Q8_0

Best Quality

—

Accuracy Retained

Calculate VRAM Hugging Face Compare

Quantization Variants

Per-quant VRAM, quality loss, and inference speed on RTX 4090

Measured = run on this site · Estimated = not run here: calculated, or a figure this site has not verified · Community = public reports

Format	Level	BPW	VRAM	PPL Loss	Speed	Source	Actions
GGUF	Q4_K_S	4.59	20.9 GB	—	—	Estimated	Calc HF
GGUF	Q4_K_M	4.88	22.2 GB	—	—	Estimated	Calc HF
GGUF	Q5_K_M	5.7	25.9 GB	—	—	Estimated	Calc HF
GGUF	Q8_0	8.51	38.6 GB	—	—	Estimated	Calc HF

Similar models

Compare with Qwen3-VL 30B-A3B

30B-A3B

Qwen3-VL 30B-A3B Instruct

Alibaba Qwen3-VL

Consumer GPUPro GPU

15.2 GBmin VRAM99.7%accuracy

Multimodal MoE with only ~3B active parameters, so it stays responsive on Apple unified memory and survives CPU offload far better than a dense 30B. Q4 ~19GB — a 24GB card holds it outright.

26B-A4B

Gemma 4 26B-A4B IT

Google Gemma 4

Consumer GPUPro GPU

15.3 GBmin VRAM—accuracy

Google's Gemma 4 mixture-of-experts model: about 25B parameters in the text model, roughly 4B active per token, a 256K window and image input. Only 5 of its 30 layers keep a cache that grows with context, and those use fewer, wider heads, so the cache stays small — about 0.9 GB at 32K and 5.3 GB at the full 256K. At Q4 that is about 17 GB at 32K: comfortable on a 24 GB card, which can still load the whole window (tight). A 16 GB card is just too small. Sizes use the calculator's generic rates — no GGUF file size could be checked, and no per-level quality loss has been published.

31B

Gemma 4 31B IT

Google Gemma 4

Consumer GPUPro GPU

18.6 GBmin VRAM—accuracy

Google's dense Gemma 4 flagship: about 31B parameters, a 256K window and image input. Only 10 of its 60 layers cache the full context, and those use four wide KV heads, so 128K of context adds about 11 GB of cache where a conventional 60-layer design would need about 120 GB. At Q4 it is about 21 GB at 4K — on a 24 GB card that is right at the edge of comfortable, and tight from 32K. A 32 GB card holds 32K comfortably. Sizes use the calculator's generic rates — no GGUF file size could be checked, and no per-level quality loss has been published.

35B-A3B

Qwen3.6 35B-A3B

Alibaba Qwen3.6

Consumer GPUPro GPU

23.4 GBmin VRAM—accuracy

One of the most recommended models for a 24 GB card this year: 35.6B parameters, about 3B active per token, so it reads per token about as much as a 3B model. Only 10 of its 40 layers keep a growing KV cache, so context is cheap: at Q4_K_M it needs about 23.4 GB at 4K and 24.0 GB at 32K. On this site's rule that is tight on a 24 GB card, not comfortable, because it counts a working buffer most guides leave out. A 32 GB card, or a 48 GB Mac through the GPU's share of unified memory, has real room. It reads images through a separate vision encoder. Sizes come from the published GGUF file sizes, and no per-level quality loss has been published.

Running MiniCPM-V 4.7 35B-A3B locally

At Q4_K_M and 4K of context, MiniCPM-V 4.7 35B-A3B needs about 22.2 GB — 20.1 GB of weights, 0.08 GB of KV cache and a 2.0 GB activation buffer. The smallest cards in this index that clear that comfortably are 32 GB ones — 4 of them, from the RTX 5090 up — and 26 of the 78 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.

What longer context costs

Going from 4K to 32K adds about 0.55 GB, taking the total to 22.8 GB. That is gentler than the parameter count suggests: only 10 of 40 layers keep a KV cache that grows with context, so the cache scales at a fraction of the usual rate. The model's native window is 256K; holding all of it at Q4_K_M would need about 28 GB.

Which build to download

This index only tracks MiniCPM-V 4.7 35B-A3B in GGUF, across 4 levels (Q4_K_S, Q4_K_M, Q5_K_M, Q8_0). No per-level perplexity sweep has been published for this model, so the quality column is empty rather than estimated — within one model, more bits per weight is the only ordering the data supports. No throughput has been measured for this model on this site's hardware. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 — CUDA only, and now archived, so local use rather than serving.

Common questions

How much VRAM does MiniCPM-V 4.7 35B-A3B need?: About 22.2 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 22.8 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.
Will MiniCPM-V 4.7 35B-A3B run on a 32GB GPU?: Yes — at Q4_K_M and 4K context it needs about 22.2 GB, which leaves 9.8 GB spare on a RTX 5090, or on any of the other 32 GB cards here (4 in all). That is the smallest size in this index that clears it comfortably; 26 of 78 do.
Which quantization of MiniCPM-V 4.7 35B-A3B should I use?: No published quality comparison exists for this model, so pick by footprint: Q4_K_M is the level most people run, and a higher bits-per-weight level is more faithful. All 4 levels in the index are Q4_K_S, Q4_K_M, Q5_K_M, Q8_0. Note this is a mixture-of-experts model — all parameters must be resident even though only a fraction are active per token, so the memory cost follows the total, not the active count.

Where this model fits

Sized at Q4_K_M with a 4K context window, smallest card first. Comfortable means the estimate uses at most 88% of the memory.

Runs comfortably on: RTX 509032GB · +9.8 Arc Pro B70 32G32GB · +9.8 Radeon AI PRO R9700 32G32GB · +9.8 Instinct MI100 32G32GB · +9.8 Mac M3 Pro 36G36GB · +4.8
Tight but possible: RTX 309024GB Radeon RX 7900 XTX24GB RTX 409024GB
Just misses: Radeon RX 7900 XT+2.2GB over Mac M4 24G+4.2GB over
Ships in: GGUF
A top pick for: Best local LLM for 24GB
Try it yourself: Size it in the calculator Generate the run command

How to actually run this

Deployment guides for this model and this class of hardware.

This page's figures change when the model or the runtime does.Last updated 2026-10-20 RSS → /feed.xml