Mistral
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.
At a glance
- Input / 1M
- $0.20
- Output / 1M
- $0.60
- Context
- 33K
- Max output
- 26K
- Intelligence
- 6.5
- Released
- Feb 2025
Pricing
Saba API pricing
Per-token rates, plus what common tasks actually cost.
| Input tokensPer 1M tokens | $0.20 |
|---|---|
| Output tokensPer 1M tokens | $0.60 |
| Cached input (read)Per 1M tokens | $0.02 |
| Cache writePer 1M tokens | Not available |
Chat message
2K in, 500 out
$0.70
per 1,000 requests
Coding task
30K in, 4K out
$8.40
per 1,000 requests
Long document summary
150K in, 2K out
$31.20
per 1,000 requests
Benchmarks
Saba benchmark scores
Independent scores from Artificial Analysis. Higher is better.
Intelligence index
6.5
Coding index
Not available
Agentic index
Not available
Specs
Context window and capabilities
What it can read, what it can write, and which API features it supports.
- Context window
- 33K tokens
- Max output
- 26K tokens
- Input types
- File, Text
- Output types
- Text
- Reasoning effort
- Not available
- Release date
- February 17, 2025
- Knowledge cutoff
- September 30, 2024
- Tool calling
- Structured outputs
- JSON mode
- Reasoning
- Temperature
- Stop sequences
- Deterministic seed
- Verbosity control
Compare
Saba vs other models
See how Saba stacks up head to head on price, benchmarks, and features.
VSSaba vs Mistral Small 4
Mistral vs Mistral
See comparisonVS
Saba vs Qwen3 VL 30B A3B Instruct
Mistral vs Qwen
See comparisonVSSaba vs Llama 3.3 70B Instruct
Mistral vs Meta
See comparisonVSSaba vs Qwen3 30B A3B
Mistral vs Qwen
See comparisonVSSaba vs Qwen3 VL 8B Instruct
Mistral vs Qwen
See comparisonVS
Saba vs GPT-4o-mini
Mistral vs OpenAI
See comparisonAlternatives
Saba alternatives
Similar models worth considering, and where each one has the edge.
Mistral
Mistral Small 4
Qwen
Qwen3 VL 30B A3B Instruct
Meta
Llama 3.3 70B Instruct
Want to learn about other models? Browse all models
FAQ
Frequently asked questions
How much does Saba cost?
Saba costs $0.20 per 1M input tokens and $0.60 per 1M output tokens through the API. A typical chat message costs about $0.70 per 1,000 messages.
What is the context window of Saba?
Saba accepts up to 33K tokens of input and can write up to 26K tokens in a single response.
What is the best alternative to Saba?
Mistral Small 4 from Mistral is a strong alternative. It offers: higher intelligence score, larger 262k context, accepts image.
Can I use Saba alongside other models?
Yes. Shortcut Chat sends one prompt to Saba and any other models you pick at the same time, so you can compare answers side by side.
Try Saba next to every other model.
Use Saba alongside all the other models. Use one at a time, or multiple together. Get the best answer from every AI.