MI

MiniCPM3 4B

MiniCPMGGUF by QuantFactoryUpdated 2024-10-08
Download on Hugging Face
Parameters4.07B
Context32K
Layers62
Architectureminicpm3
Smallest quant1.6 GB
Quick answer

MiniCPM3 4B needs about 10.8 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 14 quants, from 1.6 GB (Q2_K) to 4.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 31.2 GB.

MiniCPM3 4B – 4.07B parameter text model with 32k context

MiniCPM3 4B is a 4.07B parameter model with a 32k token context window. It supports function call and code interpreter. The model is released under apache-2.0 and is available in multiple GGUF quantizations.

Good at

  • Function calling
  • Code interpretation
  • Long‑context reasoning

Limits

  • No vision support
  • Requires 8k context for memory estimates
  • License is apache-2.0

Best for

  • Chatbots needing long context
  • Code generation tools
  • Educational tutoring

Who it’s for. Developers building local LLM applications that need a balance of size and performance. Researchers testing 4B models with extended context windows.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K1.6 GB 9.9 GB
Too big Get file →
Q3_K_S1.8 GB 10.1 GB
Too big Get file →
Q3_K_M2.0 GB 10.3 GB
Too big Get file →
Q3_K_L2.2 GB 10.5 GB
Too big Get file →
Q4_02.3 GB 10.6 GB
Too big Get file →
Q4_K_S2.4 GB 10.7 GB
Too big Get file →
Q4_K_M2.5 GB 10.8 GB
Too big Get file →
Q4_12.6 GB 10.9 GB
Too big Get file →
Q5_02.8 GB 11.1 GB
Too big Get file →
Q5_K_S2.8 GB 11.1 GB
Too big Get file →
Q5_K_M2.9 GB 11.2 GB
Too big Get file →
Q5_13.1 GB 11.4 GB
Too big Get file →
Q6_K3.3 GB 11.6 GB
Too big Get file →
Q8_04.3 GB 12.6 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (2.34 GB file, ~10.6 GB memory). For 12 GB VRAM, Q5_1 (3.07 GB file, ~11.4 GB memory) is suitable. Higher quants like Q8_0 need 12.6 GB memory and 4.33 GB file.

What GPU do you need for MiniCPM3 4B?

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: the best quant that fits is Q6_K (3.3 GB file, about 11.6 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (4.3 GB file, about 12.6 GB in use).

Which MiniCPM3 4B quant should you choose?

Q4_K_M is the usual starting point for MiniCPM3 4B: it balances file size and quality well. Here it is a 2.5 GB file.

The Q8_0 file is 4.3 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 Q2_K at 1.6 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

Architectureminicpm3
Parameters4.07B
Layers62
Trained context32,768 tokens
KV cache (fp16)930 KB per token

Source

Quantized byQuantFactory
FamilyMiniCPM
Licenseapache-2.0
Downloads (HF)1,434
Last updated2024-10-08

Frequently asked questions about MiniCPM3 4B

How much VRAM does MiniCPM3 4B need?

About 10.8 GB at Q4_K_M with an 8K context window. That is the 2.5 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 MiniCPM3 4B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 9.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of MiniCPM3 4B GGUF?

It depends on the quant: Q2_K is 1.6 GB, Q3_K_S is 1.8 GB, Q3_K_M is 2.0 GB, Q3_K_L is 2.2 GB, Q4_0 is 2.3 GB, Q4_K_S is 2.4 GB and 8 more.

What context length does MiniCPM3 4B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 7.8 GB at 8K and 31.2 GB at 32K.

Where can I download MiniCPM3 4B GGUF?

The files are on Hugging Face in the QuantFactory/MiniCPM3-4B-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 MiniCPM3 4B 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 MiniCPM3 4B on 8 GB VRAM?

Yes, using Q4_0 or Q4_K_M quantizations which require about 10.6–10.8 GB memory at 8k context.

Does MiniCPM3 4B support image input?

No, the model does not have vision capabilities.

What license applies to MiniCPM3 4B?

The model is released under the apache-2.0 license.

Similar models

Browse all →

More from QuantFactory

All by QuantFactory →

More MiniCPM models

All MiniCPM →

Other small models

All small models →