MiniCPM V 4.7 35B A3B CRACKv5 2p0 needs about 23.7 GB of memory at Q4_K_M (a 22.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 4 quants, from 0.8 GB (BF16) to 34.3 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 CRACKv5 2p0 – a 35B‑parameter multimodal MoE model in GGUF format
MiniCPM‑V 4.7 35B‑A3B CRACKv5 2p0 is a 35‑billion‑parameter multimodal model based on the qwen35moe architecture. It supports vision and runs on a variety of platforms including iOS, Android, HarmonyOS, and several chip families. The model is released under the Apache‑2.0 license.
Good at
- Vision‑enabled multimodal inference
- Mixture‑of‑experts efficient on‑device performance
- Supports multiple quantization schemes
Limits
- Requires 35B parameters; large VRAM for full precision
- No text‑only mode specified
- License is Apache‑2.0
Best for
- On‑device multimodal inference
- Mobile and edge deployments
- Developers needing a MoE model
Who it’s for. Developers and researchers building multimodal applications on mobile or edge devices. The model is suitable for those who need vision capabilities and can handle the memory requirements of a 35B‑parameter MoE model.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| BF16 | 0.8 GB | 2.0 GB | Fits | Get file → | |
| IQ2_M | 16.9 GB | 18.1 GB | Too big | Get file → | |
| Q4_K_M | 22.6 GB | 23.7 GB | Too big | Get file → | |
| Q8_0 | 34.3 GB | 35.5 GB | Too big | Get file → |
For 8 GB VRAM, use the Q4_K_M quant (22.55 GB file, ~23.7 GB memory at 8k context). For 16 GB VRAM, the IQ2_M quant (16.94 GB file, ~18.1 GB memory) is a good balance. The BF16 quant (0.79 GB file) is suitable for systems with very limited VRAM but will need more CPU memory.
What GPU do you need for MiniCPM V 4.7 35B A3B CRACKv5 2p0?
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: the best quant that fits is BF16 (0.8 GB file, about 2.0 GB in use).
- 6 GB VRAM: the best quant that fits is BF16 (0.8 GB file, about 2.0 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (0.8 GB file, about 2.0 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (0.8 GB file, about 2.0 GB in use).
- 16 GB VRAM: the best quant that fits is BF16 (0.8 GB file, about 2.0 GB in use).
- 24 GB VRAM: the best quant that fits is Q4_K_M (22.6 GB file, about 23.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q4_K_M (22.6 GB file, about 23.7 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (34.3 GB file, about 35.5 GB in use).
Which MiniCPM V 4.7 35B A3B CRACKv5 2p0 quant should you choose?
Q4_K_M is the usual starting point for MiniCPM V 4.7 35B A3B CRACKv5 2p0: it balances file size and quality well. Here it is a 22.6 GB file.
The Q8_0 file is 34.3 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is BF16 at 0.8 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about MiniCPM V 4.7 35B A3B CRACKv5 2p0
How much VRAM does MiniCPM V 4.7 35B A3B CRACKv5 2p0 need?
About 23.7 GB at Q4_K_M with an 8K context window. That is the 22.6 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 CRACKv5 2p0 on an 8 GB GPU?
Yes, with BF16: the 0.8 GB file needs about 2.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MiniCPM V 4.7 35B A3B CRACKv5 2p0 GGUF?
It depends on the quant: BF16 is 0.8 GB, IQ2_M is 16.9 GB, Q4_K_M is 22.6 GB, Q8_0 is 34.3 GB.
What context length does MiniCPM V 4.7 35B A3B CRACKv5 2p0 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 CRACKv5 2p0 GGUF?
The files are on Hugging Face in the dealignai/MiniCPM-V-4.7-35B-A3B-CRACK-GGUF 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 CRACKv5 2p0 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 this model on 8 GB VRAM?
Yes, using the Q4_K_M quantization which requires about 23.7 GB of VRAM at 8k context, but this exceeds 8 GB. A smaller quant like IQ2_M or BF16 would be needed for 8 GB VRAM.
Does this model support vision?
Yes, the model has detected vision abilities and supports single‑image, multi‑image, and video understanding.
What license does it use?
The model is released under the Apache‑2.0 license.