Mistral AI/Mixture of Experts

Mistral Small 4 119B

Name: Mistral Small 4 119B
Author: Mistral AI

Mistral Small 4 — 119B MoE with vision, coding, math. Very efficient.

chatcodingreasoningmathvisiontool_useThinkingTool Use

119B

Parameters (6.5B active)

250K

Context length

Benchmarks

Quantizations

200K

HF downloads

Architecture

MoE

Released

2026-03-16

Layers

KV Heads

Head Dim

128

Family

mistral

Quantization Options

Quant	Bits	VRAM	Quality
IQ2_XXS	2.38	35.9 GB	low
IQ2_M	2.93	44.1 GB	low
Q2_K	3.16	47.5 GB	low
IQ3_XXS	3.25	48.8 GB	low
IQ3_XS	3.5	52.6 GB	low
Q3_K_S	3.64	54.6 GB	low
IQ3_M	3.76	56.4 GB	low
Q3_K_M	4	60.0 GB	low
Q3_K_L	4.3	64.5 GB	moderate
IQ4_XS	4.46	66.8 GB	moderate
Q4_K_S	4.67	70.0 GB	moderate
Q4_K_M	4.89	73.2 GB	good
Q5_K_S	5.57	83.3 GB	good
Q5_K_M	5.7	85.3 GB	good
Q6_K	6.56	98.1 GB	excellent
Q8_0	8.5	126.9 GB	lossless
FP16	16	238.5 GB	lossless

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

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN MISTRAL SMALL 4 119B NOW

Rent on RunPod →Or Vast.ai →

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

Community Ratings

Loading ratings...

Benchmarks (11)

Arena Elo1480

IFEval84.0

AIME83.8

GPQA Diamond71.2

LiveCodeBench63.6

BBH52.7

MMLU-PRO50.7

MATH49.5

GPQA24.9

MUSR17.2

HLE4.3

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run mistral:119b-q4_K_M

Tag may need adjustment — check ollama.com/library/mistral for available tags.

▸ 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

HuggingFace GGUF Downloads Build Hardware