Microsoft-backed Mistral develops AI reasoning model for European languages: report

Mistral, a French artificial intelligence startup backed by Microsoft (NASDAQ:MSFT), plans to release a new reasoning model today, Magsitral, which would compete with similar reasoning models, such as OpenAI’s o1/o3, Anthropic’s Claude 3.7 Thinking and DeepSeek-R1 (DEEPSEEK), according to CNBC.

“We’re announcing in a couple of hours our new reasoning model, which is very much competitive with all the others and has the specificity of being able to reason in multiple languages,” Mistral CEO Arthur Mensch said to CNBC during a fireside chat at London Tech Week.

This is Mistral’s first released reasoning model, and it will initially be launched in European languages, as Mensch said most of the current reasoning models are offered in English or Chinese, according to the report. Magistral is also adept at coding and mathematics.

Reasoning models are a type of specialized language models.

“They are designed to break down complex problems into smaller, manageable steps and solve them through explicit logical reasoning,” according to Microsoft. “Unlike general-purpose LLMs which might generate direct answers, reasoning models are specifically trained to show their work and follow a more structured thought process.”

Earlier this month, Apple (AAPL) released a research paper called “The Illusion of Thinking,” which identified limitations in the abilities of large reasoning models.

“By comparing LRMs with their standard LLM counterparts under equivalent inference compute, we identify three performance regimes: (1) low-complexity tasks where standard models surprisingly outperform LRMs, (2) medium-complexity tasks where additional thinking in LRMs demonstrates advantage, and (3) high-complexity tasks where both models experience complete collapse,” the paper said. “We found that LRMs have limitations in exact computation: they fail to use explicit algorithms and reason inconsistently across puzzles.”

Seeking Alpha reached out to Mistral for comment.