Saba pricing & price history
Saba by Mistral AI costs $0.20 per 1M input tokens and $0.60 per 1M output tokens, with a 33K token context window.
Price history
No price changes since tracking began on 2026-09-24. We check hourly.
Cost calculator
Cheaper alternatives to Saba
| Model | Input | Output | Context | Intel. | |
|---|---|---|---|---|---|
N Nemotron 3.5 Lightning NVIDIA | $0.08 | $0.20 | 262K | 12.9 | compare |
D DeepSeek V4 Flash 0731 DeepSeek | $0.04 | $0.32 | 1.3M | 34.3 | compare |
M Ministral 3 8B 2512 Mistral AI | $0.15 | $0.15 | 262K | 5.5 | compare |
M Llama 4 Scout Meta Llama | $0.10 | $0.30 | 1.3M | — | compare |
X MiMo-V2.6-Flash Xiaomi | $0.14 | $0.28 | 1M | — | compare |
M Llama Guard 4 12B Meta Llama | $0.18 | $0.18 | 164K | — | compare |
O GPT-6 Luna Pro OpenAI | $0.10 | $0.50 | 1.1M | — | compare |
O GPT-6 Luna OpenAI | $0.10 | $0.50 | 1.1M | 37.3 | compare |
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...
FAQ
How much does Saba cost?
Saba by Mistral AI costs $0.20 per 1M input tokens and $0.60 per 1M output tokens, with a 33K token context window. A typical request with 2,000 input and 500 output tokens costs about $0.0007.
What is the context window of Saba?
Saba supports up to 32,768 tokens of context and up to 26,214 output tokens per request.
Has Saba’s price changed?
No price changes recorded since ModelBank started tracking it on 2026-09-24.
Is Saba being deprecated?
No retirement date has been announced for Saba as of 2026-09-24.
What are cheaper alternatives to Saba?
Cheaper options include Nemotron 3.5 Lightning ($0.08/$0.20), DeepSeek V4 Flash 0731 ($0.04/$0.32), Ministral 3 8B 2512 ($0.15/$0.15), Llama 4 Scout ($0.10/$0.30).