Mistral: Saba
by Mistral AItext+file->text3 endpoints
Overview
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...
Capabilities
Text generation
Tool use / Function calling
Structured output
Prompt caching
File input
Top-P sampling
Stop sequences
Deterministic seed
Modalities
Input
TextFile
Output
Text
Technical Specifications
Context Window
32.8K tokens
Max Output
26.2K tokens
Knowledge Cutoff
2024-09-30
Tokenizer
Mistral
Uptime (24h)
10000.0%
Supported Parameters
frequency_penaltymax_tokenspresence_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_p
Pricing
Per 1M tokens
Input$0.20
Output$0.60
Cache read$0.02
Added
February 17, 2025
Last synced 9/20/2026