Saba overview
Saba specs
- Model ID
mistralai/mistral-saba- Developer
- Mistral AI
- Context window
- 32,768 tokens
- Input
- Image, Text
- Output
- Text
- API
- OpenAI-compatible, base URL
https://api.aimlapi.com/v1
Coding and code review
Generate, debug, and review code across languages, reading whole files or repos in a single prompt.
Agentic and multi-step workflows
Reliable tool-calling and instruction-following suit agent loops that plan, call tools, and iterate.
Complex reasoning
Extended reasoning improves reliability on multi-step math, logic, and technical analysis.
Long-document analysis
Summarize and extract from contracts, papers, and reports without aggressive chunking.
Use cases
Conversational support: Mistral Saba is documented as ideal for scenarios requiring swift and precise Arabic responses, including virtual assistants that engage users in natural, real-time conversations in Arabic.
Domain-specific expertise: The model is described as useful when fine-tuned into specialized experts for fields such as energy, financial markets, and healthcare, with responses grounded in Arabic language and culture.
Cultural content creation: Mistral Saba is documented as suited to generating culturally relevant content, including educational resources, and to creating authentic materials that resonate with Middle Eastern audiences.
Translation and language support: The model is described as handling translation and language tasks across Arabic and several related regional languages while preserving cultural context and nuance.
Capabilities
- Function calling
- Reasoning
- Streaming
- Structured outputs
- Web search
Saba benchmarks by 
Artificial Analysis
See how Saba scores on independent evaluations. Data retrieved Nov 3, 2026.
Benchmark results
Compare 11 models available on AI/ML API across 4 independent benchmarks.
Humanity's Last Exam
Expert-written questions across academic subjects.
GPQA Diamond
Graduate-level science questions written to resist web search.
MMLU-Pro
Multiple-choice knowledge and reasoning across 14 subjects.
MATH-500
Competition mathematics problems.
Saba API pricing
| Usage | Price | Unit |
|---|---|---|
| Input | $0.2116 | 1M tokens |
| Output | $0.6348 | 1M tokens |
| Cached input | $0.0212 | 1M tokens |
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Saba vs other models: token prices
AI/ML API input and output rates for 5 models, per million tokens.
| Model | Input | Output |
|---|---|---|
 Saba | $0.2116 | $0.6348 |
 Command R (08-2024) | $0.1587 | $0.6348 |
| Small 4 | $0.1984 | $0.7935 |
| Mistral Large 4 | $1.36 | $4.18 |
| Medium 3.5 | $1.59 | $7.93 |
How to use the Saba API
View docs
Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/chat/completions' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "mistralai/mistral-saba",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "mistralai/mistral-saba",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}
response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "mistralai/mistral-saba",
"messages": [
{
"role": "user",
"content": "Write a haiku about the ocean."
}
]
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());Models similar to Saba
Related articles
Saba FAQ
What is the context window of Saba?
Saba has a 32,768 tokens context window.
What inputs and outputs does Saba support?
Saba takes image, text as input and returns text.
What is the model ID for Saba?
Use mistralai/mistral-saba as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
How much does Saba cost?
Saba costs $0.2116 input and $0.6348 output per 1M tokens on AI/ML API, billed as you go with no subscription.
Does Saba support streaming?
Yes, Saba can stream responses as they are generated.
Can Saba read images?
Yes, Saba accepts image input alongside text.
Who created Saba?
Saba was built by Mistral AI.
What is Saba by Mistral AI?
Saba is a chat model from Mistral AI accessible via the endpoint api.aimlapi.com/v1/chat/completions, supporting text and image inputs with text output.
Does Saba support function calling and tool use?
Yes, Saba supports function calling, parallel tool calls, and structured outputs.