MI

MiniCPM V 4.7 35B A3B

MiniCPMGGUF by Harsha001Updated 2026-10-24
Download on Hugging Face
Parameters34.7B
Context256K
Layers40
Architectureqwen35moe
Smallest quant11.2 GB
Quick answer

MiniCPM V 4.7 35B A3B needs about 19.8 GB of memory at Q4_0 (a 18.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 8 quants, from 11.2 GB (Q2_0) to 36.4 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.

MiniCPM V 4.7 35B A3B – a vision‑language MoE model

MiniCPM V 4.7 35B A3B is a 34.7B parameter model with a 3B active expert set. It supports reasoning, tools, and vision. The model is licensed under apache‑2.0 and trained on 262144 context tokens.

Good at

  • Reasoning
  • Tools
  • Vision

Limits

  • Mixture of experts requires more memory
  • No vision support in ROCmFPX quants
  • Large context token limit may need high VRAM

Best for

  • On‑device inference with low memory
  • Vision‑language tasks on AMD GPUs
  • Text‑only inference with ROCmFPX

Who it’s for. Developers building vision‑language applications on AMD hardware. Researchers needing a MoE model with efficient on‑device performance.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_011.2 GB 12.4 GB
Too big Get file →
Q4_018.5 GB 19.7 GB
Too big Get file →
Q4_018.6 GB 19.8 GB
Too big Get file →
Q4_018.7 GB 19.8 GB
Too big Get file →
Q4_019.7 GB 20.9 GB
Too big Get file →
Q4_022.7 GB 23.9 GB
Too big Get file →
Q6_K28.3 GB 29.5 GB
Too big Get file →
Q8_036.4 GB 37.6 GB
Too big Get file →

For 8 GB VRAM use Q4_0_ROCMFP4 (file size 18.5 GB, estimated 19.7 GB memory at 8k context). For tighter budgets use Q2_0 (11.18 GB, 12.4 GB memory). Higher precision quants such as Q6_K or Q8_0 require 29.5 GB and 37.6 GB memory respectively.

What GPU do you need for MiniCPM V 4.7 35B A3B?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q2_0 (11.2 GB file, about 12.4 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_0 (22.7 GB file, about 23.9 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (28.3 GB file, about 29.5 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (36.4 GB file, about 37.6 GB in use).

Which MiniCPM V 4.7 35B A3B quant should you choose?

Q4_0 (18.7 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is Q2_0 at 11.2 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters34.7B (approximate, mixture of experts)
Active parameters3B
Layers40
Trained context262,144 tokens
KV cache (fp16)80 KB per token

Source

Quantized byHarsha001
FamilyMiniCPM
Licenseapache-2.0
Downloads (HF)1,011
Last updated2026-10-24

Frequently asked questions about MiniCPM V 4.7 35B A3B

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

About 19.8 GB at Q4_0 with an 8K context window. That is the 18.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run MiniCPM V 4.7 35B A3B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiniCPM V 4.7 35B A3B GGUF?

It depends on the quant: Q2_0 is 11.2 GB, Q4_0 is 18.5 GB, Q4_0 is 18.6 GB, Q4_0 is 18.7 GB, Q4_0 is 19.7 GB, Q4_0 is 22.7 GB and 2 more.

What context length does MiniCPM V 4.7 35B A3B support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.

Where can I download MiniCPM V 4.7 35B A3B GGUF?

The files are on Hugging Face in the Harsha001/MiniCPM-V-4.7-35B-A3B-ROCmFPX repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does MiniCPM V 4.7 35B A3B use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run MiniCPM V 4.7 35B A3B on 8 GB VRAM?

Use the Q4_0_ROCMFP4 quant; it estimates 19.7 GB memory at 8k context, so 8 GB VRAM is insufficient. Lower‑bit quants are needed.

Does the ROCmFPX fork support vision input?

No, the ROCmFPX quants are text‑only. Vision input requires upstream llama.cpp with standard quants.

What is the active parameter count?

The model has 3.0B active parameters in the MoE configuration.

Other GGUFs of MiniCPM-V-4.7-35B-A3B

Similar models

Browse all →

More from Harsha001

All by Harsha001 →

More MiniCPM models

All MiniCPM →

Other large models

All large models →