Mistral: Saba is an API-access large language model from Mistralai priced at $0.20/1M input tokens and $0.60/1M output tokens with a 33k context window. Context caching reduces input costs to $0.10/1M for cache reads.
Economics (OpenRouter)
Input / 1M
$0.20
Output / 1M
$0.60
Est. Monthly Cost
$10.26
Based on current usage settings
Cache Read$0.10
Cache Write$0.25
Capabilities (models.dev)
ReasoningNo
Tool CallNo
Audio InNo
Audio OutNo
Modalitytext
Context Window33k
Benchmarks (AA)
Intelligence Index
General reasoning and knowledge
—
Coding Index
Code generation & debugging
—
Math Index
Mathematical reasoning
—
Performance
—tokens/sec
* Median output throughput
Model ID: mistralai/mistral-saba • Data Sources: OpenRouter, Artificial Analysis, models.dev