MI

MiniCPM V 4.7 1B DSpark

MiniCPMGGUF by openbmbUpdated 2026-10-20
Download on Hugging Face
Parameters214M
Context256K
Layers5
Architecturedflash
Smallest quant0.1 GB
Quick answer

MiniCPM V 4.7 1B DSpark needs about 0.7 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 11 quants, from 0.1 GB (Q4_0) to 0.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.3 GB.

MiniCPM V 4.7 1B DSpark – a lightweight vision‑enabled draft model

MiniCPM V 4.7 1B DSpark is a 0.214 billion‑parameter draft checkpoint built on the dflash architecture. It supports vision and tool use and is designed to pair with the full MiniCPM‑V‑4.7‑1B model. The model is released under the Apache‑2.0 license.

Good at

  • Vision‑enabled text generation
  • Tool‑aware responses
  • Draft acceptance with minimal memory

Limits

  • No autonomous intent; outputs may be inaccurate
  • Requires dflash architecture and DSpark Markov head
  • No official benchmarks provided

Best for

  • Rapid prototyping of vision‑enabled LLMs
  • Testing draft‑based speculative decoding
  • Low‑memory inference on 8‑GB GPUs

Who it’s for. Developers and researchers who need a small, vision‑capable draft model for speculative decoding or rapid experimentation. The model is suitable for local deployment with llama.cpp and for integration into multimodal pipelines.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_00.1 GB 0.7 GB
Fits Get file →
Q4_K_S0.1 GB 0.7 GB
Fits Get file →
Q4_K_M0.1 GB 0.7 GB
Fits Get file →
Q4_10.1 GB 0.7 GB
Fits Get file →
Q5_00.2 GB 0.7 GB
Fits Get file →
Q5_K_S0.2 GB 0.7 GB
Fits Get file →
Q5_K_M0.2 GB 0.7 GB
Fits Get file →
Q5_10.2 GB 0.8 GB
Fits Get file →
Q6_K0.2 GB 0.8 GB
Fits Get file →
Q8_00.2 GB 0.8 GB
Fits Get file →
F160.4 GB 1.0 GB
Fits Get file →

For 8 GB VRAM, the Q4_0, Q4_K_S, or Q4_K_M quantizations use about 0.7 GB of VRAM at 8 k context and are the smallest options. If more precision is needed, Q5_0 or Q5_K_S use about 0.8 GB. The Q8_0 quant uses 0.8 GB and is a good default for stability.

What GPU do you need for MiniCPM V 4.7 1B DSpark?

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 F16 (0.4 GB file, about 1.0 GB in use).

Which MiniCPM V 4.7 1B DSpark quant should you choose?

Q4_K_M is the usual starting point for MiniCPM V 4.7 1B DSpark: it balances file size and quality well. Here it is a 0.1 GB file.

The Q8_0 file is 0.2 GB, about 1.8 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 Q4_0 at 0.1 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedflash
Parameters214M
Layers5
Trained context262,144 tokens
KV cache (fp16)10 KB per token

Source

Quantized byopenbmb
FamilyMiniCPM
Licenseapache-2.0
Downloads (HF)14,687
Last updated2026-10-20

Frequently asked questions about MiniCPM V 4.7 1B DSpark

How much VRAM does MiniCPM V 4.7 1B DSpark need?

About 0.7 GB at Q4_K_M with an 8K context window. That is the 0.1 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 1B DSpark on an 8 GB GPU?

Yes, with F16: the 0.4 GB file needs about 1.0 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 1B DSpark GGUF?

It depends on the quant: Q4_0 is 0.1 GB, Q4_K_S is 0.1 GB, Q4_K_M is 0.1 GB, Q4_1 is 0.1 GB, Q5_0 is 0.2 GB, Q5_K_S is 0.2 GB and 5 more.

What context length does MiniCPM V 4.7 1B DSpark support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.3 GB at 32K.

Where can I download MiniCPM V 4.7 1B DSpark GGUF?

The files are on Hugging Face in the openbmb/MiniCPM-V-4.7-1B-DSpark-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 1B DSpark 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 1B DSpark on 8 GB VRAM?

Yes, the Q4_0, Q4_K_S, and Q4_K_M quantizations require roughly 0.7 GB of VRAM at 8 k context, fitting comfortably on an 8 GB GPU.

Does the model support vision inputs?

Yes, the model includes vision capabilities and can process image prompts when paired with the appropriate vision projector.

What is the recommended quant for best performance?

The Q8_0 quantization offers the highest precision and is a solid default, but Q4_0 or Q4_K_S provide a good balance between size and speed for most use cases.

Other GGUFs of MiniCPM-V-4.7-1B-DSpark

Similar models

Browse all →

More from openbmb

All by openbmb →

More MiniCPM models

All MiniCPM →

Other tiny models

All tiny models →