Mistral
Ministral 3 3B (3B parameters) requires approximately 5.6 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 7 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run Ministral 3 3B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "mistralai/Ministral-3-3B-Instruct-2512" \
--hf-file "Ministral-3-3B-Instruct-2512-Q4_K_M.gguf" \
-c 4096 -ngl 99Quick specs
About this model
Related models
Quick picks
Best hardware
Run this model
Quantization options
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.2 GB | Low | — |
Q3_K_S | 3 | 1.5 GB | Low | — |
NVFP4 | 4 | 1.7 GB | Medium | — |
Q4_K_M | 4 | 1.8 GB | Medium | — |
Q5_K_M | 5 | 2.2 GB | High | — |
Q6_K | 6 | 2.5 GB | High | — |
Q8_0 | 8 | 3.2 GB | Very High | — |
F16 | 16 | 6.1 GB | Maximum | — |
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
Ministral 3 3B (3B parameters) requires approximately 5.6 GB of VRAM with Q4_K_M quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, Intel Arc A380 6GB can run Ministral 3 3B with a compatibility score of 74/100. It provides 6 GB of memory and achieves approximately 42.0 tokens per second.
The recommended quantization for Ministral 3 3B is Q4_K_M, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for Ministral 3 3B: RTX 3050 8GB (score: 77/100), RTX 3060 Ti 8GB (score: 77/100), RTX 3070 8GB (score: 77/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, Ministral 3 3B is well-suited for chat as well as vision, agentic. It was designed with these use cases in mind.
See also
