Skip to content
LLMs
Mistral AI logo

Mistral: Saba

by Mistral AI

Mistral: Saba is a large language model from Mistral AI. It costs $0.200 per million input tokens and $0.600 per million output tokens. Its context window is 33K tokens.

Tool callingStructured outputPrompt cachingfile input
Input / 1M tokens
$0.200
Output / 1M tokens
$0.600
Cached input / 1M
$0.020

On repeated prefixes

Context window
33K

tokens

Who serves it cheapest

1 hosts serve Saba. Same weights, same API - the price difference is pure margin and routing.

Providers serving Mistral: Saba, cheapest first
ProviderInput / 1MOutput / 1MContextThroughputUptime 24h
MistralCheapest$0.200$0.60033K-100.0%

About Saba

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...

Specifications

Mistral: Saba specifications
Model IDmistralai/mistral-saba
ProviderMistral AI
Context window33K tokens
Max output26K tokens
Input modalitiestext, file
Output modalitiestext
Knowledge cutoff2024-09-30
Open weightsNo
ReleasedFebruary 17, 2025

Cheaper alternatives

Models that cost less than Saba while keeping at least half its context window and every input modality it supports.

Frequently asked

How much does Mistral: Saba cost?

$0.200 per million input tokens and $0.600 per million output tokens. Cached input reads cost $0.020 per million tokens.

What is the context window of Mistral: Saba?

33K tokens, with up to 26K tokens of output per request.

Confirm against the source: Mistral AI official pricing.