FitMyLLM
Mistral AI/Dense

Mistral AIMinistral 3 8B

Ministral 3 8B — the 8B of Mistral's Ministral 3 line: dense, 256K context, tool use, and the size most single-GPU setups settle on.

chatcodingmultilingualtool_use
8.92B
Parameters
256K
Context length
15
Benchmarks
6
Quantizations
132K
HF downloads
Architecture
Dense
Released
2025-12-02
Layers
34
KV Heads
8
Head Dim
128
Family
mistral

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q4_K_M4.89
7.5 GB
5.9 + 1.6 KV
good
Q5_K_S5.57
8.3 GB
6.7 + 1.6 KV
good
Q5_K_M5.7
8.4 GB
6.8 + 1.6 KV
good
Q6_K6.56
9.4 GB
7.8 + 1.6 KV
excellent
Q8_08.5
11.6 GB
10.0 + 1.6 KV
lossless
FP1616
19.9 GB
18.3 + 1.6 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
Ready to run this?Rent by the hour

Rent a GPU and run Ministral 3 8B now

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (15)

Run this model

Easiest way to get started·Beginners
Docs ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run ministral-3:8b-instruct-2512-q4_K_M

Downloads and runs automatically. Add --verbose for speed stats.

Setup guide
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA Tesla C2070
6 GB VRAM • 143 GB/s
NVIDIA
NVIDIA Tesla C2075
6 GB VRAM • 150 GB/s
NVIDIA
NVIDIA Tesla C2090
6 GB VRAM • 177 GB/s
NVIDIA
NVIDIA Tesla M2070
6 GB VRAM • 150 GB/s
NVIDIA
NVIDIA Tesla M2070-Q
6 GB VRAM • 150 GB/s
NVIDIA
NVIDIA Tesla M2075
6 GB VRAM • 150 GB/s
NVIDIA
NVIDIA Tesla M2090
6 GB VRAM • 177 GB/s
NVIDIA
NVIDIA Tesla X2070
6 GB VRAM • 177 GB/s
NVIDIA
NVIDIA Tesla X2090
6 GB VRAM • 177 GB/s
NVIDIA
NVIDIA Tesla K20X
6 GB VRAM • 250 GB/s
NVIDIA
NVIDIA Tesla K20Xm
6 GB VRAM • 250 GB/s
NVIDIA

Find the best GPU for Ministral 3 8B

Build Hardware for Ministral 3 8B
Spec sheet

Ministral 3 8B — 8.92B Dense.

Specifications
Parameters
8.92B
Architecture
Dense Transformer
Context Length
256K tokens
Capabilities
chat, coding, multilingual, tool_use
Release Date
2025-12-02
Provider
Mistral AI
Family
mistral
VRAM requirements
QuantBPWVRAMQuality
Q4_K_M4.895.9 GB94%
Q5_K_S5.576.7 GB96%
Q5_K_M5.76.8 GB96%
Q6_K6.567.8 GB97%
Q8_08.510.0 GB100%
FP161618.3 GB100%
Benchmark scores
MMLU-PRO64.2
GPQA Diamond47.1
LiveCodeBench30.3
AIME31.7
HLE4.3
AA Intelligence9.0
AA Coding9.7
AA Math31.7
IFBench29.1
Terminal-Bench Hard4.5
τ²-Bench26.6
SciCode20.8
AA Long Context25.3
Terminal-Bench 2.14.1
Terminal-Bench 40.0
Run command

Run Ministral 3 8B locally with Ollama — needs 5.9 GB VRAM at Q4_K_M:

$ollama run ministral-3:8b
Compatible GPUs
30 @ Q4_K_M