Ministral 3 14B
Ministral 3 14B loads in 9.1 GB at Q4_K_M, comfortably from 24 GB of memory, Macs included, starting at the Mac Mini M6 16GB. Below is the memory math we ran for it, the smallest card that fits, and the fastest machine we track.
- Parameters
- 13.5B
- Format
- Q4_K_M
- Min memory
- 16 GB
- Best for
- Chat, Reasoning, Vision
How much memory does Ministral 3 14B need?
Ministral 3 14B is a 13.5B dense model built for chat, reasoning, vision. The weights alone take 9.1 GB at Q4_K_M; add context and the OS, and it belongs in the 24 GB memory class. Every number below comes from the ModelFit engine: speeds are estimates, labeled est., because we do not benchmark hardware ourselves.
KV = fp16 estimate (q8_0 cache roughly halves it). "Comfortable" = weights + KV within the engine's tiered budget (~70-85% of memory).
Which machines run Ministral 3 14B?
- Smallest GPU
- NVIDIA GeForce RTX 3060est. 30 tok/s · 12 GB VRAMtypical error 15%
- Fastest
- NVIDIA GeForce RTX 5090est. 142 tok/stypical error 15%
- Runs on a Mac?
- Mac Mini M6 16GBest. 15 tok/stypical error 10%
Verdicts come from the same engine that powers our fit checker: memory capacity first, bandwidth for speed. How we compute these numbers · What typical error means · Hugging Face model card
More Mistral models
12 pagesGet told when a better model fits your MacBook Air M5 16 GB
Get one email when a new open-weight model is a better fit for this machine.
For: MacBook Air M5 16 GB
You'll also get the weekly ModelFit email. By signing up you agree to our Privacy Policy. Unsubscribe from either email anytime, in one click.
Config reports
No speed test run for this model yet. Run the speed test: npx @modelfit/cli@latest bench
Share your config
Frequently asked questions
How much memory does Ministral 3 14B need?
9.1 GB for the Q4_K_M weights, plus ~3.0 GB of KV-cache at 16k context, about 12.1 GB total. Comfortable from 24 GB of VRAM or unified memory.
Does Ministral 3 14B run on a Mac?
Yes, from the Mac Mini M6 16GB (est. 15 tok/s, typical error 10%). Unified memory means the RAM budget is the only limit.
What is the smallest GPU for Ministral 3 14B?
The NVIDIA GeForce RTX 3060 (12 GB VRAM) is the smallest tracked card that runs Ministral 3 14B comfortably, est. 30 tok/s, typical error 15%.
Cite this page
ModelFit: Ministral 3 14B, specs, memory math and hardware verdicts. https://modelfit.io/models/ministral-3-14b/ (dataset updated 2026-11-16, CC BY 4.0).