mistral
14B parameters
Commercial OK
Multimodal
Reviewed July 2026

Ministral 3 14B

Ministral 3 14B is Mistral AI's dense 14B from December 2025, released under Apache-2.0 with vision input and a 256K context. The Q4_K_M build is 9.1GB, which targets 16GB cards once KV cache is counted.

License: Apache-2.0·Context: 262,144 tokens

Overview

Ministral 3 14B is Mistral AI's dense 14B from December 2025, released under Apache-2.0 with vision input and a 256K context. The Q4_K_M build is 9.1GB, which targets 16GB cards once KV cache is counted.

Strengths

  • Apache-2.0 license with commercial use allowed
  • Vision input plus 256K context in a single dense build
  • 14B quality tier while staying under 10GB of weights at Q4_K_M

Weaknesses

  • 9.1GB weights plus long-context KV cache is tight on 12GB cards
  • Dense 14B decodes slower than 9B peers on the same GPU
BLK · QUALITY BENCHMARKreviewed · raw logs

Reviewed quality benchmarks

First-party rows were run by RunLocalAI; reviewed community rows are labeled in the data. Every row links to the raw test-run log.

BenchmarkQuantRuntime / HardwareScoreRaw log
HumanEval+
tested 2026-07-20
Q4_K_M
ollama-0.32
rtx-3080-16gb-mobile
76.8/100
Gist →

Q4_K_M note:First-party humaneval+ on RTX 3080 Laptop 16GB via Ollama 0.32. Windows-safe scoring. July 2026 campaign.

Want to verify? Every row links to its Gist with full stdout and stderr of the run. The runner script is in the public repo (scripts/run-humaneval-plus.ts) — reproducible end-to-end. Browse all coding scores at /benchmarks/coding.

Quantization variants

Each quantization trades model quality for file size and VRAM. Q4_K_M is the most popular starting point.

QuantizationFile sizeVRAM required
Q4_K_M9.1 GB12 GB

Get the model

Ollama

One-line install

ollama run ministral-3:14bRead our Ollama review →

Hardware that runs this

Cards with enough VRAM for at least one quantization of Ministral 3 14B.

NVIDIA GB200 NVL72
13824GB · nvidia
AMD Instinct MI350X
288GB · amd
NVIDIA B300 (Blackwell Ultra)
288GB · nvidia
AMD Instinct MI355X
288GB · amd
AMD Instinct MI325X
256GB · amd
AMD Instinct MI300X
192GB · amd
NVIDIA B200
192GB · nvidia
NVIDIA H100 NVL
188GB · nvidia
Compare alternatives

Models worth comparing

Same parameter band, plus what's one tier above and below — so you can decide what actually fits your hardware.

Frequently asked

What's the minimum VRAM to run Ministral 3 14B?

12GB of VRAM is enough to run Ministral 3 14B at the Q4_K_M quantization (file size 9.1 GB). Higher-quality quantizations need more.

Can I use Ministral 3 14B commercially?

Yes — Ministral 3 14B ships under the Apache-2.0, which permits commercial use. Always read the license text before deployment.

What's the context length of Ministral 3 14B?

Ministral 3 14B supports a context window of 262,144 tokens (about 262K).

How do I install Ministral 3 14B with Ollama?

Run `ollama pull ministral-3:14b` to download, then `ollama run ministral-3:14b` to start a chat session. The default quantization is Q4_K_M.

Does Ministral 3 14B support images?

Yes — Ministral 3 14B is multimodal and accepts text + vision inputs. Vision support requires a runner that handles its image-conditioning architecture.

Source: Vendor official documentation

Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify model claims.

Related — keep moving

Compare hardware
Buyer guides
When it doesn't work
Recommended hardware
Before you buy

Verify Ministral 3 14B runs on your specific hardware before committing money.

Will it run on my hardware? →Custom hardware comparison →GPU recommender (4 questions) →