Mistral AI launches Magistral Small 1.1 – powerful AI inference on RTX 4090
Mistral AI is releasing Magistral Small 1.1, a 24-billion parameter model capable of running locally on a single Nvidia RTX 4090 graphics card, introducing open-access reasoning steps.
What happened?
Mistral AI has launched Magistral Small 1.1, a compact 24-billion parameter AI model designed for local inference. The model is optimised to run locally on a single consumer graphics card, specifically the Nvidia RTX 4090. A key technical innovation is the introduction of an open reasoning mechanism via dedicated [THINK] tags, which display the model's chain of thought, making the decision-making process transparent and auditable.
Key facts
| Modellstorlek | 24 miljarder parametrar |
|---|---|
| Filstorlek (GPTQ) | 12 GB |
| Hårdvarukrav | 1x Nvidia RTX 4090 |
| Lanseringsdatum | 1 oktober 2026 |
Why it matters
The model addresses the conflict between performance, cost, and data integrity that has affected many companies throughout 2025 and 2026. By running advanced reasoning models locally on standard hardware, businesses can drastically reduce API expenses while keeping sensitive data within their own networks. The open reasoning feature also increases traceability in industries requiring auditability and oversight.
Who is affected?
The launch is primarily aimed at small and medium-sized enterprises, developers, and organisations in sensitive sectors such as finance and healthcare. It is designed for users who require high-level reasoning capabilities in AI applications but wish to avoid high API costs or cannot transmit sensitive data to cloud services.
Impact on the EU
The model is distributed under an open license, making it accessible to European companies and developers. Its capability for local execution facilitates compliance with strict EU regulations such as the GDPR and the requirements of the EU AI Act regarding data integrity and transparency.
What else you should know
Using GPTQ quantisation, the model's size is compressed to 12 GB, preserving 92 per cent of the precision compared to the FP16 format while fitting onto consumer hardware. In Mistral's own tests, the transparency feature using [THINK] tags was shown to increase user trust in the model's conclusions by 62 per cent.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Hur påverkas användare i EU?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Mistral AI launches Magistral Small 1.1 – powerful AI infere"