Magistral Small 1.2
- Next shutdown
- None setNo date on 2 platforms
- GPQA Diamond
- 47.6%#93 of 108 · Epoch AI
- OTIS Mock AIME 2024-2025
- 28.1%#85 of 105 · Epoch AI
- Context window
- 128Ktokens
- Released
- Sep 18, 2025Mistral AI
Dates on every platform
2 listings · a floor is the earliest possible date
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Amazon BedrockAmazon Web Services: Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab) · checked Oct 14, 2026 | mistral.magistral-small-2509Active | – | None set | – |
| Mistral APIMistral AI: Models Overview (opens in a new tab) · checked Oct 14, 2026 | magistral-small-2509Shut down · also magistral-small-latest | Shut down | Mistral Small 4 |
Mistral's models overview still showed the status Deprecated on 14 October 2026, after the retirement date.
What breaks when you move off Magistral Small 1.2
Intelligence
Third-party results, credited below
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| GPQA Diamond | Epoch AI | 47.6% ± 2.6% | #93 of 108 | Aug 30, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 28.1% ± 5.6% | #85 of 105 | Aug 30, 2026 |
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
$ per 1M tokens, with each price’s source
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
Amazon Bedrockmistral.magistral-small-2509 | $0.50 | – | $1.50 | Oct 1, 2026Current · Standard on-demand tier in us-east-1. | Amazon Web Services: AWS Price List API: AmazonBedrock, us-east-1 (opens in a new tab) |
Change history
ShutdownMistral Small 3.2 and Magistral Small 1.2 retire from the Mistral APIMistral API1 break
mistral-small-2506 and magistral-small-2509 retired on 31 July 2026. Mistral names Mistral Small 4 as the replacement for both.
- Platform
- Mistral API
- Replacement
- Mistral Small 4
- Announced
- BehaviorMagistral models always returned thinking traces. Mistral Small 4 and Mistral Medium 3.5 reason through the reasoning_effort parameter: high returns a thinking chunk before the answer, none omits it.
Source: Mistral AI, Models Overview (opens in a new tab); Mistral AI, Native reasoning (deprecated) (opens in a new tab); Mistral AI, Reasoning (opens in a new tab)