MI

Ministral 3 3B Base 2512

MistralGGUF by mradermacherUpdated 2025-12-04
Download on Hugging Face
Parameters3.43B
Context256K
Layers26
Architecturemistral3
Smallest quant0.0 GB
Quick answer

Ministral 3 3B Base 2512 needs about 3.5 GB of memory at Q4_K_M (a 2.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 25 quants, from 0.0 GB to 2.8 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 3.5 GB.

Ministral 3 3B Base 2512 – a compact vision‑enabled language model

Ministral 3 3B Base 2512 is a 3.43B‑parameter model with a 0.4B vision encoder. It supports a 256k context window and is licensed under Apache 2.0. The base version is not fine‑tuned for instruction or reasoning tasks.

Good at

  • Vision‑enabled text generation
  • Multilingual support
  • Edge deployment

Limits

  • Not instruction‑tuned
  • Large context window may need more VRAM
  • Requires 16GB VRAM in BF16 or less with quantization

Best for

  • Image captioning
  • Real‑time translation
  • Fine‑tuning for specialized tasks

Who it’s for. Developers building lightweight, real‑time AI on edge or low‑resource devices. The model is suitable for custom post‑training pipelines and local deployment.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 1.4 GB
Fits Get file →
IQ1_S0.9 GB 2.3 GB
Fits Get file →
IQ1_M1.0 GB 2.4 GB
Fits Get file →
IQ2_XXS1.1 GB 2.5 GB
Fits Get file →
IQ2_XS1.2 GB 2.6 GB
Fits Get file →
IQ2_S1.2 GB 2.6 GB
Fits Get file →
IQ2_M1.3 GB 2.7 GB
Fits Get file →
Q2_K_S1.4 GB 2.7 GB
Fits Get file →
IQ3_XXS1.4 GB 2.8 GB
Fits Get file →
Q2_K1.5 GB 2.8 GB
Fits Get file →
IQ3_XS1.6 GB 3.0 GB
Fits Get file →
Q3_K_S1.6 GB 3.0 GB
Fits Get file →
IQ3_S1.7 GB 3.0 GB
Fits Get file →
IQ3_M1.7 GB 3.1 GB
Fits Get file →
Q3_K_M1.8 GB 3.2 GB
Fits Get file →
Q3_K_L1.9 GB 3.3 GB
Fits Get file →
IQ4_XS2.0 GB 3.3 GB
Fits Get file →
Q4_02.0 GB 3.4 GB
Fits Get file →
IQ4_NL2.1 GB 3.4 GB
Fits Get file →
Q4_K_S2.1 GB 3.4 GB
Fits Get file →
Q4_K_M2.1 GB 3.5 GB
Fits Get file →
Q4_12.2 GB 3.6 GB
Fits Get file →
Q5_K_S2.4 GB 3.8 GB
Fits Get file →
Q5_K_M2.5 GB 3.8 GB
Fits Get file →
Q6_K2.8 GB 4.2 GB
Fits Get file →

For 8GB VRAM, use Q4_K_M (2.15 GB file, ~3.5 GB memory at 8k context). For 4GB VRAM, use Q3_K_M (1.8 GB file, ~3.2 GB memory). For 2GB VRAM, use IQ2_XS (1.19 GB file, ~2.6 GB memory).

What GPU do you need for Ministral 3 3B Base 2512?

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 Q5_K_M (2.5 GB file, about 3.8 GB in use).
  • 6 GB VRAM: the best quant that fits is Q6_K (2.8 GB file, about 4.2 GB in use).

Which Ministral 3 3B Base 2512 quant should you choose?

Q4_K_M is the usual starting point for Ministral 3 3B Base 2512: it balances file size and quality well. Here it is a 2.1 GB file.

If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemistral3
Parameters3.43B
Layers26
Trained context262,144 tokens
KV cache (fp16)104 KB per token

Source

Quantized bymradermacher
FamilyMistral
Licenseapache-2.0
Downloads (HF)699
Last updated2025-12-04

Frequently asked questions about Ministral 3 3B Base 2512

How much VRAM does Ministral 3 3B Base 2512 need?

About 3.5 GB at Q4_K_M with an 8K context window. That is the 2.1 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 Ministral 3 3B Base 2512 on an 8 GB GPU?

Yes, with Q6_K: the 2.8 GB file needs about 4.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Ministral 3 3B Base 2512 GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 0.9 GB, IQ1_M is 1.0 GB, IQ2_XXS is 1.1 GB, IQ2_XS is 1.2 GB, IQ2_S is 1.2 GB and 19 more.

What context length does Ministral 3 3B Base 2512 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.5 GB at 32K.

Where can I download Ministral 3 3B Base 2512 GGUF?

The files are on Hugging Face in the mradermacher/Ministral-3-3B-Base-2512-i1-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 Ministral 3 3B Base 2512 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 8GB VRAM?

Yes, using Q4_K_M or Q4_K_S quantizations will keep memory usage below 8GB at 8k context.

Does this model support vision tasks?

Yes, it includes a 0.4B vision encoder for image analysis and captioning.

Is the model fine‑tuned for instruction following?

No, the base model is not fine‑tuned; use the Instruct variant for chat or instruction use cases.

Other GGUFs of Ministral-3-3B-Base-2512

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Mistral models

All Mistral →

Other tiny models

All tiny models →