Do you want smarter insights in your inbox? Sign up for our weekly newsletters to get what is crucial for AI leaders, data and security. Subscribe now
From Openai introduced deep research, an AI agent who can Perform tests for users and generate a comprehensive report, many other companies have published THEIR Own versions of this ability, all found deep research.
Deep research, as a function and product, can be obtained through various platforms, including Google’s GeminiIN AlfasenseIN Ty.comIN DeepseekIN GROK 3 And many others. Now the French company Mistral He joins the fight against the launch of deep research options in his chat, including platform updates.
IN Blog postThe company said that deep research and other recent features will make Le chat “even more talented, intuitive and more fun.”
LE chat users can open the research mode and ask something. Chatbot then asks questions about the explanation of some information, and then begins the collection of sources. He will submit a “structured, supported by a reference report, which is easy to follow.”
The AI Impact series returns to San Francisco – August 5
The next AI phase is here – are you ready? Join the leaders from Block, GSK and SAP to see the exclusive look at how autonomous agents transform the flows of the work of the company-decision-making in real time for comprehensive automation.
Secure your place now – the space is restricted: https://bit.ly/3guplf
Mistral said that his research is powered by a deep research agent, which he designed as “really helpful” and seems to be cooperation with an organized research partner.
Deep research has been called “the first artificial mass intelligence that can displace jobs”, especially since they can issue reports faster than human analysts.
Mistral also updated the “thinking mode”, in which LE chat users can access the company’s chain model to read and answer in different languages. It can also switch the code in the middle of the sentence.
Edition of images based on quick and other functions
For people creating images in the chat application, they can ask Chatbot to edit part of the photo using hints. Users can say something like “generate a cat drawing”, and then ask Le Chat to “put it in Istanbul” and he will do it.
Thanks to the recently released model of speech recognition Voxtral, Le Chat Can Export Four Mode, in which users can talk to the platform loudly. The company said that this mode is the best to recognize low -delay speech and keep up with someone’s conversational pace.
The recent function of Le Chat projects allows users to organize related conversations and topics in groups. The projects will utilize their own libraries – which may contain sent files – and keep the tools and preferred settings. This is similar to Google notebook.
Playing up
Many recent functions on Le Chat may seem familiar. Other chat platforms are normal to introduce similar functions, especially when people begin to expect these opportunities when using chat systems.
For example, both Gemini and CHATGPT allow users to edit generated photos using a poem. Sometimes chatbots misunderstand and process the whole image. Recently, however, I generated and edited the photo from chatgpt, and Chatbot removed exactly what I wanted.
Voice mode has been available at ChatGPT since September 2024, although Chatgpt has always contained the “Read out” function. Project Astra from Google has gone voice mode to a recent level, showing in a demo that users can indicate something in the physical world to the twins and ask for it aloud.
However, Mistral has the advantage that there is Europe and can introduce functions directly to the European market. Companies such as OPENAI often try to bring certain services, such as the advanced CHATGPT voice mode, to Europe because of the provisions regarding data and certain provisions of the AI of the European Union.
Despite this, users seemed excited that Mistral brought many recent powerful functions of Le Chat, and some early users see good results from Mistral’s deep research.
