All AI models · LLM benchmarks · Methodology

Mistral

Saba

Mistral model for multilingual chat, reasoning, and tool-assisted workflows

Model facts

Context window
33K tokens
Maximum output
26K tokens
Cheapest paid input
$0.199 per 1M tokens
Cheapest paid output
$0.595 per 1M tokens
Open weights
yes
Providers
3

Capabilities and modalities

tool calling, structured output, attachments, temperature control, open weights, text, pdf.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
default6.2not measurednot measurednot measured

Observed price history

  • 2026-08-24: $0.199 input / $0.595 output per 1M tokens

API providers

  • NanoGPT (Mistral) — model id mistralai/mistral-saba; $0.199 input / $0.595 output per 1M tokens; provider documentation
  • OpenRouter — model id mistralai/mistral-saba; $0.2 input / $0.6 output per 1M tokens; provider documentation
  • Kilo Gateway (Mistral) — model id mistralai/mistral-saba; $0.2 input / $0.6 output per 1M tokens; provider documentation