Prototype · closed beta
unOpen.AI
In closed beta: Roxy, a private AI assistant with permanent memory running on infrastructure the user owns. Brand, site, product, backend and conversational design.
Mistral · Mistral AI
Saba.
Saba is a Mistral AI model, listed on 17 February 2025. It takes text and files and returns text, with a 32,768-token context window.
- Hosted API
- 33k context
Specification
- Provider
- Mistral AI
- Family
- Mistral
- Listed
- 17 February 2025
- Catalogue id
- mistralai/mistral-saba
- Context window
- 32,768 tokens
- Maximum output
- 26,214 tokens
- Inputs and output
- Text and files in; text out
- Capabilities
- Tool calling and structured outputs
- Weights
- Not published — hosted API
- Knowledge cut-off
- 2024-09-30
- Listed price
- $0.200 input, $0.600 output, $0.020 cached input per million tokens
Provider-published specifications as of 30 October 2026, not figures we measured. Model families move quickly — check the provider's own documentation before committing to a number.
Source: Saba in the OpenRouter catalogue, checked weekly. Confirm current terms in Mistral AI's own documentation before you commit.
What it costs
Price the whole answer.
Illustrative calculation at listed rates: a batch of 10 million input tokens and 1 million output tokens costs $2.60 on Saba. A thousand requests of 2,000 input and 500 output tokens each costs $0.700. Calculation: input millions × input rate + output millions × output rate; it excludes reasoning tokens, retries, tools, tax and review time.
Compare cost per accepted result, not per token: replay the same work on two or three candidates and count the answers you would actually ship. Our model-routing approach sends each request to the cheapest model that passes.
THE REST OF THE FAMILY
Other Mistral models.
Every Mistral model in the catalogue. Our Mistral page covers where the family fits and how we evaluate it.
| Saba | 17 Feb 2025 | — | — | 33k | $0.200 | $0.600 |
|---|---|---|---|---|---|---|
| Mistral Large 4 open | 29 Oct 2026 | 26.0 | 169 | 1.05M | $0.700 | $2.10 |
| Mistral Medium 3.5 | 29 Apr 2026 | — | 152 | 262k | $1.50 | $7.50 |
| Mistral Small 4 open | 16 Mar 2026 | 4.3 | 170 | 262k | $0.150 | $0.600 |
| Devstral 2 2512 open | 9 Dec 2025 | — | — | 262k | $0.400 | $2.00 |
| Ministral 3 14B 2512 open | 2 Dec 2025 | — | — | 262k | $0.200 | $0.200 |
| Ministral 3 3B 2512 open | 2 Dec 2025 | — | — | 131k | $0.100 | $0.100 |
| Ministral 3 8B 2512 open | 2 Dec 2025 | — | — | 262k | $0.150 | $0.150 |
| Mistral Large 3 2512 | 1 Dec 2025 | — | — | 262k | $0.500 | $1.50 |
| Voxtral Small 24B 2507 open | 30 Oct 2025 | — | — | 33k | $0.100 | $0.300 |
| Mistral Medium 3.1 | 12 Aug 2025 | — | — | 131k | $0.400 | $2.00 |
| Codestral 2508 | 1 Aug 2025 | — | — | 256k | $0.300 | $0.900 |
| Mistral Small 3.2 24B open | 20 Jun 2025 | — | — | 256k | $0.075 | $0.200 |
| Mistral Medium 3 | 7 May 2025 | — | — | 131k | $0.400 | $2.00 |
| Mistral Small 3.1 24B open | 17 Mar 2025 | — | — | 128k | $0.351 | $0.555 |
Intelligence and coding: Artificial Analysis indices at the model's best reasoning setting, and its measured output speed. Context and prices: the OpenRouter catalogue list price in USD per million tokens. Released: the provider's date where Artificial Analysis records one, otherwise the date the catalogue first listed it. Refreshed weekly; not measured by QQuantum.ai. Select a heading to sort.
Questions
Questions about Saba.
When was Saba listed?
Saba was listed on 17 February 2025 in the OpenRouter catalogue; the provider's own announcement may predate it.
What is the context window of Saba?
32,768 tokens, with up to 26,214 output tokens per response, as listed in the OpenRouter catalogue on 30 October 2026. A long window is not the same as reliable recall across it; evaluate retrieval on your own documents.
How much does Saba cost?
The catalogue lists $0.200 per million input tokens and $0.600 per million output tokens, and $0.020 per million cached input tokens (checked 30 October 2026). Mistral AI's own rates, batch discounts and cloud marketplaces can differ.
Can Saba run on our own hardware?
The catalogue lists no published weights for it, so it is used as a hosted API. Where data may not leave your infrastructure, an open-weight model is the alternative.
