crawler-status-logs-v1
AI API Hunt
M

Mistral Saba

Last updated: Recently

Model Type

MText

Max Context

32.8K

Max Output

-

Parameter Scale

-

Mistral Saba is 24B parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. 32,768 token context window. Includes independent benchmarks from Artificial Analysis.

Input Modality

text, file

Output Modality

text

Inference Speed

-

Latest Release Date

2/17/2025

SDK Ecosystem

-

artificial-analysismanufactureraa-bootstrap
Model Overview Fields

name

Mistral Saba

Release date

2/17/2025

Performance

-

model_creator.name

Mistral

AA Evaluation Scores

AA Intelligence Index

12.1

AA Coding Index

-

AA Math Index

-

MMLU Pro

0.611

GPQA

0.424

HLE

0.041

LiveCodeBench

-

SciCode

0.241

Math-500

0.676666666666667

AIME

0.13

AIME 2025

-

IFBench

-

LCR

-

TerminalBench Hard

-

TAU2

-

Output Speed (tokens/s)

-

TTFT (s)

-

First Answer Token (s)

-

Mistral logo

Mistral

Mistral Saba logo

Mistral Saba

$0.00/M$0.00/M

Mistral Saba is 24B parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. 32,768 token context window. Includes independent benchmarks from Artificial Analysis.

00
0.0
Share:
Report / Feedback