Mistral: Saba

Mistral·Released 2025-02-17

Open Source33K ctx

About

Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional...

Quality Index

12.1

390th of 507

Top 78%

Price/1M

$0.30

235th cheapest

46% below median

Top 38%

Context Window

33K

353rd largest

Top 92%

Pricing

Input

$0.20

per 1M tokens

Output

$0.60

per 1M tokens

Blended

$0.30

per 1M tokens

Cheaper than 62% of models. Median price is $0.56/1M tokens.

Cost Calculator

Tokens per day1M

100K100M

Daily

$0.30

Monthly

$9.00

vs. Similar Models

AllenAI: Olmo 3 32B ThinkQ:0.0

$0.24-21%

Olmo 3.1 32B InstructQ:+0.1

$0.300%

Qwen: Qwen-TurboQ:-0.1

$0.06-81%

Reka Flash (Sep '24)Q:-0.1

$0.35+17%

Performance

Context Window

33K

tokens

Larger than 8% of models

Context Window Comparison

DeepSeek: R1 Distill Qwen 32B

33KSame

Qwen2.5 Coder 7B Instruct

33KSame

Mixtral 8x7B Instruct

33KSame

Benchmarks

MMLU-Pro

61.1%

GPQA Diamond

42.4%

HLE

4.1%

LiveCodeBenchNot evaluated

SciCode

24.1%

TerminalBench HardNot evaluated

MATH-500

67.7%

AIME

13.0%

AIME 2025Not evaluated

IFBenchNot evaluated

Long Context RecallNot evaluated

Tau2Not evaluated

Market AverageTop Score

Open Source

Quick Compare

Similar Models

AllenAI: Olmo 3 32B Think

Allen AI

Q: 12.1$0.24/1M66K ctx

Cheaper: 21%Context Window: 2x larger

DeepSeek R1 Distill Llama 8B

DeepSeek

Q: 12.1N/A/1M

Gemma 4 E2B (Non-reasoning)

Google

Q: 12.1N/A/1M

Olmo 3.1 32B Instruct

Allen AI

Q: 12.2$0.30/1M66K ctx

Context Window: 2x larger

Qwen: Qwen-Turbo

Alibaba

Q: 12.0$0.06/1M131K ctx

Cheaper: 81%Context Window: 4x larger

R1 1776

Perplexity

Q: 12.0N/A/1M

Compare all 7 models

Quality Index

12.1

390th of 507

Top 78%

Price/1M

$0.30

235th cheapest

46% below median

Top 38%

Context Window

33K

353rd largest

Top 92%