Mistral AI Introduces Magistral Reasoning Models
Mistral AI has announced the release of its first reasoning models, Magistral Small and Magistral Medium, announced in a press release. These models are designed to excel in domain-specific, transparent, and multilingual reasoning, offering both open-source and enterprise solutions.
Magistral Small, a 24 billion parameter model, is available for download under the Apache 2.0 license, allowing the community to modify and build upon its architecture. Magistral Medium, a more powerful version, is available in preview on Mistral's Le Chat platform and through various cloud services.
The models are fine-tuned for multi-step logic, providing a traceable thought process in multiple languages, including English, French, Spanish, and more. Magistral Medium is noted for its speed, delivering answers up to 10 times faster than competitors in Le Chat, and is suited for a wide range of enterprise applications, from legal research to software development.
Magistral's release is supported by comprehensive evaluations and a focus on transparent reasoning, making it a significant contribution to AI research and development. The open-source nature of Magistral Small encourages community engagement and innovation in the field of reasoning language models.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Mistral AI Releases Timestral Models for Time Series Forecasting
Mistral AI Acquires New Material and Forms Solutions Design Team
Liquid AI Releases LFM2.5-24B-A2B for Local AI Agents
Overmind Open Sources Platform for Specialized AI Models
Mistral Opens Munich Hub for Industrial AI
Daily AI Brief: the AI news that matters, in your inbox.