Mistral

Europe cannot be a “technological vassal of the United States”, and the CEO of Mistral is clear about the path

by

Mistral is emerging as the pillar of European artificial intelligence. A few weeks ago we said that the French startup had raised another 830 million dollars to create AI data centers in Europe. Arthur Mensch is the CEO of the company and, for some time now, he is establishing himself as one of the powerful voices within the initiative of European technological sovereignty. His new warning is that Europe cannot be a “vassal state” of the United States and he has published a roadmap so that Europe leads AI. It won’t be easy. European swerve. There are those who complain that everything cannot be politics, but really politics is something that permeates many layers and the European turn in search of technological sovereignty has a lot to do with this. It is something that has coincided with the return of Donald Trump because Europe has realized that, between threats and the “I invaded Greenland”, can’t trust his ally. With American technology companies very involved in the ideology set by their Government, there is a demand for sovereign European alternatives that do not depend on American Big Tech nor how they may process your sensitive data. What happens with rockets, satellites, chips and even with Microsoft Office. And AI is no exception. Measures. That’s right where it comes into play. Mistral. As the greatest exponent of European AI (within the Generative AIsince we also have the suite from the Spanish Freepik as one of the most important companies in this field), Mistral and its CEO are voices with a certain weight when it comes to talking about what seems to be the only topic of conversation in recent months. And Mensch has clear that Europe cannot be a “vassal” of US technology companies. For this reason, they have published “European AI: a roadmap to lead it”, a long document in which it urges the institutions of the European Union to speed up procedures and permits to take advantage of its single market position of more than 450 million people and combine the talent of different countries at the service of AI. From European AI, of course. The premises are clear: Attract and retain talent. Unlock the full potential of the single market. Embrace European AI on all economic fronts. Empower Europe with critical infrastructure for AI. 80%. Each of the measures has a series of points that detail what the optimal way to proceed would be (according to Mistral) to achieve European leadership and stop depending on foreign technology. And one of the points to keep in mind is that Europe has the possibility of commanding, but it faces a devastating fact: 80% of the digital infrastructure continues to depend on non-EU providers. To put it down: if a ministry needs an office suite, turn to Google or Microsoft. If a hospital needs an AI, goes to ChatGPT or Huawei. If we limit ourselves to AI, Mistral estimates that only 20% of EU companies have adopted AI and that only 11% of SMEs take advantage of its potential. slap on the wrist to regulatory Europe. What they point out is that this situation makes us vulnerable to extraterritorial controls that put the strategic autonomy of the member countries in check. They defend that this roadmap is not a theoretical exercise, but rather something practical that is based on three key principles: Action over theory. Unity against the fragmentation of each of the governments. And the most important: the speed is questionable. We must find quick ways to attract talent and capital so that the most innovative in Europe are not left behind, trapped in regulations that take time. Ambition. They warn that it is something with potential not only for Mistral, but for the entire ecosystem, an ecosystem in which Mistral is already very, very well positioned. Part of the 830 million they have raised will go to their facilities near Paris where there will be 13,800 NVIDIA GB3000 chips (You can’t become independent from NVIDIA…), but it won’t be the only one. By 2027 they hope to have a €1.2 billion facility in Sweden with 23 MW of computing capacity. In total, they hope to achieve 200 MW of capacity by the end of next year. It is very, very far from China and, above all, from the United States, but although the distance is enormous, it is an important step. The B side of the matter. Now, everything has a twist, and the twist of this enormous amount of money is that this round is not venture capital, but debt financing. The main French banks have lent this money to create data centers and, while the risk capital is not returned, the debt is, and with interest. It doesn’t matter if Mistral’s move turns out well or not, even if the AI ​​bubble bursts: the banks that have lent the money expect to receive it with the aforementioned interest regardless of how business goes. It is an added pressure for the company, but also a sign that they trust in what they are building. In Xataka | ChatGPT’s milestone is not being a good AI: it is having become one of the biggest attention grabbers in history

While in the US there is a civil war over military AI, a European one has sneaked into the French Armed Forces: Mistral

by

One of the big topics of conversation in recent weeks is the civil war in the United States. between two of the AI ​​giants and the Government itself. To sum up, Anthropic gave in his artificial intelligence Claude to the Pentagon to integrate it into all its systems along with Palantir. It is estimated that it has been a key tool for capture Nicolás Madurobut also to attack Iranbut since the US wanted to go further and Anthropic refused, OpenAI took its place. It is a very strange situation because the result could be that The US blacklists Anthropic as if it were Huawei. It would be the first time they have done that to an American company and it is something that tells us two things. The first is that governments need big technology companies and their tools. The second is that technology companies also have a lot to gain. And, while all that noise is happening in the United States, in Europe another AI company has ‘sneaked’ into its country’s security systems. We refer to a Mistral that, without making noisehas been building its portfolio of contacts in European defense systems for some time. A local AI for the security of Europe With all the spotlights pointing to ChatGPT, Claude and Chinese equivalents such as DeepSeekothers have been carving out a niche for themselves. Almost without a sound, and overnight, a French company called Mistral was building multilingual models that rivaled the American alternatives that captured all eyes. Founded in June 2023, Mistral will soon converted in the French technological jewel, but also in the Europe’s AI gem. Its managers, engineers who came from Google (deepmind) and Meta, managed to attract the attention of NVIDIA, ASML, Samsung, IBM, Salesforce or a Microsoft that invested 15 million euros and incorporated Mistral models into Azure. Mistral’s policy is to release the code of its models so that anyone can analyze, use and adapt it. According to them, it is something that will accelerate innovation and, without the muscle of the American giants, their approach is to a large model with other smaller and specialized ones. At the beginning of this year, the bombshell arrived. The Ministry of the Armed Forces of France arrive to an agreement with Mistral AI to integrate the company’s models, software and services in public entities related to the ministry. For example, the Atomic Energy Commission, the National Office of Aerospace Studies and Research, the Hydrographic and Oceanographic Service of the Navy and, also, the Armed Forces. It is the Ministry of Defense Artificial Intelligence Agency that will supervise all operations, but the idea is to “deploy Mistral tools in France’s infrastructure, ensuring full control over data and critical technologies.” From the Government, it was noted that, with these new tools, they can prepare the Armed Forces for the challenges of the future. However, the fact that Mistral is the Defense model in France may matter little to us… unless we take two things into account. The first: shortly after this agreement became known, public information on how Mistral is seeking Defense contracts outside France. They are getting more and more into this path, signing agreements with Helsinga German defense startup, and Arthur Mensch, CEO of Mistral, argued that its AI models will be “instrumental in the development of a new generation of defense systems.” Here comes the second thing to take into account. Mensch himself commented that these tools “will ensure Europe’s strategic advantage on the global stage”. This is important because Europe is now at a point where it has realized that the partners and allies of the past may not be the same as those of the future. European sovereignty At the user level we see movements like those of independence from American technologiesbut at a political level, Mistral responds to that search for European sovereignty. Instead of leaving your Defense-related systems in the hands of a foreign company, these agreements mean that it is a European company that is on the inside. And, in the end, this is not just about AI. In recent months we are seeing how Europe is moving to have a more powerful voice in terms of semiconductors, computing for artificial intelligence and even in the new space race. For a long time it has been said that others innovate and Europe legislates, but the current situation It has caused Europe to continue legislating while doing everything else. Images | Mistral AI, Amio Cajander In Xataka | The United States wants to be “sovereign” on a technological level. The problem is that everything it builds depends on other countries

Mistral is the AI ​​that is playing its cards best. Because it is taking advantage of the fever for European technological sovereignty

by

To the cheetah being silent, Mistral grows like foam. The French artificial intelligence startup claims that its revenue has multiplied by 20 over the past year, and they have achieved it with a particularly striking and effective strategy: defending and promoting European technological sovereignty. what has happened. Arthur Mensch, co-founder and CEO of Mistral, explains in Financial Times that its latest annualized revenue rate — which estimates annual revenue based on last month’s revenue — was above $400 million. A year ago that rate was only 20 million a year. Or what is the same: he has multiplied it by 20. This works. The startup based in Paris hasn’t stopped to grow since its beginnings and last year already was valued at 12,000 million euros. That figure may soon become obsolete, because the company is on track to surpass $1 billion in annual recurring revenue by the end of the year if it continues this growth. Between their alliances more striking is the one who signed with ASML in September 2025: that was when the Dutch company invested 1.3 billion euros in it. It is not making too much noise, but it continues to grow with a key component. Companies in power. Mistral is rapidly expanding the number of large enterprise clients it works with. Right now it has more than 100, and although it is not especially popular among end users – who tend to choose models from Big Tech companies in the US – the option for these European companies is increasingly clear. If they want not to depend on infrastructure and control outside Europe, they now have Mistral as a great alternative. New data centers. The firm announced this Wednesday that it will invest 1.2 billion euros in a new data center in Sweden. It is the first center of its kind that the company will build outside of France, and Mensch explained that “We are diversifying and distributing our capacity throughout Europe.” That data center will be created in collaboration with EcoDataCenter, and is expected to be operational in 2027. The choice of Sweden was easy according to Mensch, who noted that it was very attractive because the energy there was “low in carbon emissions and relatively cheap.” Partners and clients deep inside but also outside the EU. Although Mistral is postulated as the great reference in terms of this “European AI”, it also has Microsoft and NVIDIA as investors. In fact its ambition is global, but the fact of being the only major European developer of foundational LLMs It has put it in the spotlight of all European companies that seek independence from partners from the US or China. ASML, Total Energies, HSBC and governments such as France, Germany and Greece already use Mistral’s services, and 60% of their revenue comes from Europe. A perfect speech for these times. The CEO of Mistral is clear about the strategy and has arrived at the right time to apply that strategy that defends European sovereignty: “Europe has realized that its dependence on American digital services was excessive and is now at a critical point. We give them (European companies) an advantage because we provide them with models, software and computing capacity completely independent of American players.” Data centers must be from European companies. Mensch also talked about all those data centers than Big Tech will create in Europe and, of course, in Spain: “It is important that we realize that it is not so useful (for States) to deploy computing resources if you only create data centers for US hyperscalers“. Or what is the same: having AI data centers from companies like Microsoft, Google or Amazon in Europe serves the interests of these companies much more than European interests. In Xataka | Europe has begun to become technologically and militarily independent from the United States. First stop: replace Starlink

The elite of the open models spoke in Chinese. Mistral has just placed Europe at a level that not even the US managed to reach

by

Over the last year, the elite of open models for assisted programming, at least in benchmarks as SWE-Bench Verifiedhas spoken with a Chinese accent. Names like DeepSeek, Kimi either qwen They had settled into the top positions in testing and were setting the pace in complex software engineering tasks, while Europe was still searching for its position. The arrival of Devstral 2 alters that distribution. It does not displace those who were already at the top, but it places Mistral at the same level of demand and turns a European company into a real contender in a field that until now seemed reserved for others. League change: the technical leap that had been brewing for some time. During recent months, the open models developed in Europe and the United States had shown constant evolution, although still without the performance necessary to compete in the most demanding tests. The progress was evident, but there was a lack of a project capable of consolidating it at a higher level and demonstrating that this path could give results comparable to those of the sector. Devstral 2 in data: performance, size and licenses. The new Mistral model reaches 123B parameters in a dense architecture and offers an expanded context of 256K tokens, accompanied by a modified MIT license that facilitates its adoption in open environments. Its compact version, Devstral Small 2, reduces the model to 24B licensed parameters Apache 2.0. In the SWE-Bench Verified figures published by the companyDevstral 2 obtains 72.2%, a mark that places it in the most competitive section of the open models evaluated and that confirms its presence among the most advanced alternatives in the segment. It is reflected by a panorama concentrated in the upper part of the benchmark. Among the open models, DeepSeek V3.2 leads the group with 73.1%, followed by Kimi K2 Thinking with 71.3% and for proposals such as Qwen 3 Coder Plus and Minimax M2, which are around 69 points. At lower levels GLM 4.6, GPT-OSS-120B, CWM and DeepSWE appear, with more moderate results. In the closed commercial environment (proprietary models), the graph incorporates higher scores: Gemini 3 Pro reaches 76.2%, GPT 5.1 Codex Max rises to 77.9% and Claude Sonnet 4.5 scores 77.2%, all of them above the best brands registered for open models. What SWE-Bench Verified Really Measures and Why It Matters. SWE-Bench Verified is a test designed to evaluate whether a model can solve real programming tasks, not synthetic exercises. Each case presents a bug in an open source repository and requires a patch to pass the previously failed tests. The evaluation seeks to measure whether the system understands the structure of the project, identifies the cause of the problem and proposes a coherent solution. It is a useful and demanding metric, although limited to Python repositories and a specific set of situations that do not cover the full breadth of software work. From co-pilots to agents who act on the project. The arrival of Devstral 2 coincides with a broader change in the way of working with programming tools. It is no longer just about receiving suggestions in the editor, but about having agents capable of exploring an entire repository, interpreting its structure and proposing changes consistent with its real state. In this context, Vibe CLI appears, a tool that allows Devstral to analyze files, modify parts of the code and execute actions directly from the terminal, bringing these capabilities closer to the daily workflow of developers. Cost and deployment: what each type of user can do with Devstral. The model will be available for free for an initial period and will then cost $0.40 per million tokens for input and $2.00 per million for output, while the Small 2 version will be priced lower. Its deployment also makes a difference: Devstral 2 requires at least four H100-class GPUs, aimed at data centers, while Devstral Small 2 is intended to run on a single GPU and, according to Mistral documentation, the Devstral Small family can also run in CPU-only configurations, without a dedicated GPU. This variety allows both companies and individual developers to find a suitable entry point. The appearance of Devstral 2 introduces an unexpected element in a space where Chinese companies set the pace and where not even the United States, despite its leadership in artificial intelligence, had an open model in this high performance range in SWE-Bench Verified. Mistral does not displace those who were already at the top, but it does broaden the conversation and shows that Europe can compete in a field where it did not appear until now. It is a movement that does not alter the general hierarchy, although it does open a new margin for the evolution of assisted programming tools. Images | Xataka with Gemini 3 In Xataka | OpenAI and Google deny that they are going to put ads in ChatGPT and Gemini. The reality is that accounts do not come only with subscriptions

A Bugatti Mistral costs five million dollars. Launching it includes convincing the police to organize a race

by

It’s not every day that you can brand new a Bugatti Mistrala supercar valued at more than five million and that the CEO of Bugatti himself come deliver it to you in person. However, it is not so common that for this delivery, the CEO has to convince the police that it is a good idea to cut off one of Miami’s coastal roads to traffic to debut the supercar by racing between the Mistral and a custom-built sports yacht for the same owner. Although it may seem very bizarre, these things can happen when you are millionaire enough. A very particular premiere in Miami The delivery of a Bugatti Mistral is never a routine event. It’s a exclusive supercar of which only 99 units were manufactured that were they sold the same day that was put up for sale. However, when you pay five million euros for one of these exclusive jewels, the least you expect is that the CEO of Bugatti himself will come to deliver it to you in person. According to published Luxury Launchesthat’s what happened to Anthony Hsieh, a millionaire from Miami who received the exclusive unit of this supercar. The staging, far from being limited to a simple presentation in the dealer who had sold it to himincluded an unusual proposal: a race in front of the sea competing head to head with one of the exclusive yachts for sport fishing that Hsieh’s company builds. Bugatti’s CEO also joins in Mate Rimac, founder of the brand Rimac supercarscurrent CEO of Bugatti and a true speed enthusiast, did not want to miss the race and got so involved that he finally ended up offering to drive the Mistral in its race against the yacht. Obviously, the CEO wasn’t going to risk getting pulled over by the police or having the car’s owner fined, so he opted to convince Miami traffic authorities to close one of Miami’s busy coastal roads for the race, and This is how he told it on his networks social. A routine delivery for a Bugatti. Bugatti Mistral W16 engine The Bugatti Mistral uses the brand’s legendary W16 engine, an engineering gem what brand the end of an era for the brand since this is the last production model that will carry this 8-liter, 4-turbo block that delivers a power of 1,600 hp. Such a beast catapults the Mistral at a speed above 453 km/h. Her opponent was not exactly a cruising yacht. It is about the Badco 50 Gameboata boat designed for sport fishing of tuna and billfish (a large species similar to swordfish) and therefore must have agile and powerful engines that allow it to navigate at speeds of up to 44 knots. Like the Bugatti, the Badco 50 are customized to the owner’s taste with materials of the highest quality and resistance. Saying that the Badco 50 is a simple fishing boat is like saying that the Mistral is just a car. Furthermore, it so happens that the company that manufactures the Badco 50 is Bad Company Fishing Adventures, It is owned by the millionaire who bought the Mistral, so organizing this race, which as you can see in the video that was recordedis more symbolic than real, the brand sought to turn the delivery of the supercar into an unrepeatable experience for its customer. It’s not every day that the head of a supercar brand makes you luxury chauffeur in the car that has just been delivered to you and all followed by a police escort. If at this point you are still wondering who was the overall winner of the racethe answer is more than obvious: Mate Rimac, and not just by driving the car fasterbut because he took in his pocket the five million that the Bugatti Mistral costs and the absolute loyalty of a customer who will never again receive a car like Bugatti did with his Mistral. In Xataka | Bugatti has discovered that millionaires no longer want to buy luxury cars: they want to buy unique works of art Image | Bad Company Fishing Adventures

To the question of what sense it makes to compete with Google, OpenAI or Anthropic in AI, Mistral has an answer: small and local models

by

French startup Mistral AI Mistral 3 has been launcheda family of 10 open source artificial intelligence models that represent its most ambitious commitment to date. The Parisian company, which is often considered the main European hope in the development of AI, seeks to differentiate itself from the large American technology companies by betting on flexibility and deployment in all types of devices instead of raw power. Under these lines we tell you all the news. What Mistral has presented. The Mistral 3 family includes a flagship model called Mistral Large 3, with 675 billion parameters, and nine compact models grouped under the name Ministral 3 (in three sizes: 14,000, 8,000 and 3 billion parameters). All models are released under Apache 2.0 license, allowing unrestricted commercial use. The large model also has multimodal capacity, being able to process text and images. It is also multilingual, with a special emphasis on European languages. On the other hand, small models can run on devices with just 4 GB of memory, making them perfect for modest laptops, mobile phones and embedded systems without the need for an internet connection. Why strategy matters. While OpenAI, Google and Anthropic focus on increasingly powerful and closed systems with agentic capabilitiesMistral has focused on the breadth and scope of its models, efficiency and what its co-founder Guillaume Lample calls “distributed intelligence.” According to declared told VentureBeat, the company believes the future of AI is defined not by scale, but by ubiquity: models small enough to run in drones, vehicles, robots and consumer devices. The economic and practical argument. Lample explained It means that in more than 90% of cases, a small, specifically tuned model can get the job done, especially if it is trained with synthetic data for specific tasks. According to Lample, this is not only cheaper and faster, but it eliminates concerns about privacy, latency and reliability. The company also has teams that work directly with customers to analyze specific problems and fine-tune small models that perform specific tasks. This, above all, can attract companies that become frustrated when choosing the best possible model for a specific task and, if it does not perform adequately, they end up giving up. Europe is lagging behind. If we talk about innovation and technology around AI, we do not hesitate to say that Europe is leagues away of what companies in the United States and China are offering. This is why Mistral AI advocates a different approach in which it prioritizes massive deployment in devices and the flexibility of its smaller models. The capacity offered by open models can be a great asset to continue betting on these technologies. In China, for example, the open models of DeepSeek, Alibaba or Kimi are emerging widelyabove in certain tasks even competitors as large as ChatGPT. Lample explained that most leading Chinese models are exclusively text-based, with separate image processing systems. For this reason, they also want to opt for a multimodal approach. A complete ecosystem. Mistral no longer only offers language models. The company has built an entire ecosystem that includes Mistral Agents APIwith connectors for code execution, web search and image generation; Masterlyyour reasoning model; Mistral Code for programming assistance; and AI Studioan application deployment platform that also has analytical and logging capabilities. Furthermore, his assistant Le Chat It has incorporated a deep research mode, voice capabilities and a list of more than 20 enterprise integrations. Thus, in addition to its model offering, the company can provide other companies with a whole layer of personalized products and services, with the aim of being their main source of financing. Digital sovereignty. Although Mistral is often characterized as Europe’s answer to OpenAI, the company prefers to consider itself as ‘a transatlantic collaboration’. Its CEO, in fact, is in the United States, has teams on both continents and trains these models in collaboration with American teams and infrastructure. However, its positioning as a defender of European digital sovereignty has earned it strategic partnerships with the French army, the country’s employment agency, the Luxembourg government and various European public organizations. The European Commission presented in October a strategy to promote European AI tools that provide security and resilience while boosting the continent’s industrial competitiveness. Offline capabilities for democratization. The use cases that Mistral has designed for its small models include, above all, local applications, such as factory robots that use sensor data in real time and without relying on the cloud, drones in natural disasters or rescues that operate offline, and smart cars with functional AI assistants in remote areas. Lample stood out that there are billions of people without internet access but with laptops or cell phones capable of running these small models, which he considers potentially revolutionary. Additionally, by running on the device, these apps preserve the privacy of user data. Real “open source” debate. Not everyone celebrates Mistral’s approach. Some critics question his decision to opt for models’open weight‘, that is, free to access but providing less information about their code than truly “open source” models, which provide the code and training data necessary to train a model from scratch. Andreas Liesenfeld, assistant professor at Radboud University and co-founder of the European Open Source AI Index, declared to the Financial Times that data at scale is the missing key in the European AI innovation ecosystem and that Mistral does not contribute to that at all. The long-term strategic bet. Lample recognize that their models are “a little behind” the most advanced closed systems, but argued that the important thing is that “they are catching up quickly.” Time will tell if Mistral’s approach to low-cost, versatile models with local applications ends up working for them to end up positioning themselves as one of the great European bets on AI. Cover image | Mistral AI In Xataka | China already has an army of 5.8 million engineers. His new plan involves accelerating doctorates

ASML, Airbus and Mistral are planted before Brussels. They ask that the application of the law of AI and notify the risks delay

by

Europe already has its great artificial intelligence law. What is missing, according to several companies, are the concrete rules to apply it. Only one month after the first standards for the most advanced models, more than 45 large companies – among them ASML, Airbus or Mistral – enter into force – They have signed an open letter asking Brussels to “stop the clock” and postpone their entry into force two years. They point to an unrealistic calendar and the difficulty of competing with the United States or China. What exactly is EU’s artificial intelligence law? The European Union Artificial Intelligence Law entered into force on August 1, 2024after having been politically approved by the European Parliament and the Council in December 2023. It is the first comprehensive regulation of the world focused on this technology, and regulates from how the models are trained to what contexts can be used. The key is in its approach to risk levels: the greater the potential impact, more legal obligations. And what exactly Asml, Airbus, Mistral and the rest ask? They demand a pause two years before the most demanding parts of the law enter into force, especially those that affect high -risk systems and the general purpose models, whose first section is scheduled for August 2025. The reason: The standard is too complex, overlaps with other regulations and still lacks key guides for its application. ASML headquarters in Veldhoven Among those guides is the code of good practices, that had to have been published in spring and still does not be ready. Companies argue that without that document, and with this level of uncertainty, the law can become a brake for European innovation. “This situation puts at risk not only the development of European leaders, but the ability of all industries to deploy the scale required by global competition,” They warn. They also ask that regulatory quality prioritize against speed, and warn that continuing without changes would send a wrong message to the seriousness of Europe in its commitment to technological competitiveness. The names behind this initiative. The request does not arise from an isolated startup or from an informal group of companies. Behind is the EU AI Champions Initiative, a group that groups more than 60 European companies that claim to be committed to the development of a competitive AI and aligned with the EU values. Among its members are names such as ASML, Airbus, Mistral AI, Mercedes-Benz, BNP Paribas, Siemens Energy, Lufthansa, Philips or Publicis. Of course, not all members of the US Ai Champions Initiative signed the letter published this week. Images | Sigmund | Rawpixel | ASML In Xataka | After strictly regulating AI, the European Union has identified a problem: it has been too European Union

I have tried the chat, the French Mistral chatbot, and has arguments to fight with Chatgpt. Not just for being faster

by

Mistral recently launched the mobile app and the new web version of Le Chat, his AI assistant that seeks a European approach (and with a French accent) for chatbots that are already something everyday in our lives. Incidentally, that launch has been accompanied by an important update of its web platform. As a regular user of Chatgpt and ClaudeI wanted to test it in several scenarios, also comparing it with its rivals. The first impression impacts: It is deviably fastmuch more than any model we have seen so far. Of course he has even more crumb. Speed ​​changes the rules The most striking is still your speed in responding. Reach 1,000 words per second thanks to Its integration with brain processors. This, in practice, means long virtually snapshot responses. Not that GPT-4O Or 3.5 Sonnet on duty are too slow, it is a first world problem, but I think we all prefer to wait a second to wait fifteen. In my tests comparing Le Chat with Chatgpt and Claude I have noticed that difference. For example, by asking them to write a long and complete analysis about the tariff policy of the United States, Le chat did it in a couple of seconds, with a well structured text and even with appointments. His rivals offered quite similar quality responses, but it took longer. And if we want to spin fine, Claude was the one who was above Chatgpt: his answers are usually better in pure writing (apart from code, for example), he has a somewhat more refined, natural and professional style, all at the same time. In fact That’s where chat loses against Claude: The most convincing style, more similar to that of a competent human, the Anthropic chatbot wins. Chatgpt maintains a good general balance, but here does not stand out in speed or style. It is also worth commenting on something: their answers are backed by consultations to reputed media and agencies such as AFP, his countryman, with whom he maintains a collaboration agreement. In my translation and content analysis tests, the model has shown to be at the height of Chatgpt and Claude in quality, although with a more concise style. A somewhat trivial example made me glimpse that Le chat seems somewhat more assertive than chatgpt or claude. Especially the first (and sometimes also the second) sin of being too influenced by the human who talks with them. The example that said: I asked him if Xataka was a Spanish soccer club, what would it be. His answer was Barcelona. He argued reasonably well, but I retended in an incisive way: “Why not Real Madrid? Xataka is the most read technology of Spain at a distance of the second, and Real Madrid is the Spanish club with more titles.” This was your answer. Colors out, I liked it to be reaffirmed and that it was not complacent with my re -question: The choice of FC Barcelona to compare with Xataka is based on several factors that go beyond the number of titles. Here are some reasons why I opted for Barcelona instead of Real Madrid: Style and philosophy: FC Barcelona is known for its distinctive game style, often associated with the “Tiki-Taka”, which is innovative and different. Xataka, being a technology blog, also stands out for its innovative approach and its ability to adapt to new technological trends. Identity and community: FC Barcelona has a strong Catalan identity and a very united community of followers. Xataka also has a community of very compromised and passionate readers by technology, which reinforces that sense of community. Adaptability and avant -garde: Although Real Madrid is a huge club with many titles, FC Barcelona has been historically known for its ability to adapt and be at the forefront in terms of play style and talent development. This aligns with how Xataka remains at the forefront of technology. Global recognition: Both Xataka and FC Barcelona have significant global recognition in their respective fields. Although Real Madrid is also globally recognized, the way in which FC Barcelona has built its unique brand and style resonates more with the identity of Xataka. These reasons do not imply that Real Madrid is less relevant, but simply highlight how certain aspects of FC Barcelona resonate more with Xataka’s identity and approach as a technology blog. By the way, when I asked who Real Madrid would be, he gave an argument argued in his prestige, global scope, awards, etc: The country. It sounded at least reasonable, but when, already restless, I asked who Valencia would be, he told me The world. Twisting the gesture, I asked him who was then Atlético de Madrid. Answer: “Public“. End of the test. Personalized agents and automation One of Le Chat’s most interesting characteristics is his agents system, which allows Create specialized attendees invoking them with “@”. This functionality, similar to OpenAI GPTS or Claude projects, but with a different approach, allows you to automate specific tasks and create personalized workflows. Yes indeed: As with GPTs and projects, this is a feature that requires a Pro subscriptionno free plans. Image: Xataka. Agents can be configured in two ways: through the visual interface of the plataforme or through the API for developers. The interesting thing is that you can customize aspects such as: The base model (Mistral Large 2Mistral Nemo or Codestral). The “temperature” or the tone of the answers (to make them more creative or more precise). Specific behavioral instructions. Examples of use to improve your performance. Unlike ChatGPT GPTS, which are more oriented to the end user, Le Chat agents seem to be designed also thinking about business integrations and automated workflows. While Claude does not yet offer similar functionality (although Claude Pro with extended context), Mistral approach seems to be halfway between the ease of use of chatgpt and flexibility They are looking for developers. In my experience, the creation of agents is more technical than with GPTS, but also more powerful in terms of customization and … Read more

Mistral AI is the French startup that opted for efficiency before Deepseek. His future is uncertain

by

Mistral ai is the French technological jewel. The AI ​​startup has become practically the only European representative that competes with large companies and technological startups in the US or China. He also does it with an absolute focus on efficiency, which is just what is now valued as Deepseek. However, its future is complex. A promising European startup. When we talk about her for the first time in 2023, he surprised that without having any product He managed to raise 105 million euros investment Soon, yes, the fruits of such commitment would begin to appear, which also generated that the firm raised More financing rounds. Flag efficiency. One of the defining characteristics of Mistral AI was that his work was always oriented to do more with less. To seek efficiency. The startup, knowing not being able to compete with the Bigt Tech in resources, has always sought to create more compact models but with great performance. He succeeded with Large 2 123b, which was three times smaller that calls 3.1 405b but matched that model and others like Claude 3.5 or GPT-4o in some metrics. They return to the load with Small 3 24b. The startup He has just announced The availability of Small 3 24b, its new “small” model. Actually it is not so much: those with size between 1b and 14. However, it is certainly an interesting llm for one thing: it competes from you to you with flame 3.3 70b – the last of the finish line, almost three sometimes bigger – and also does it with a latency (time it takes to appear every token) much smaller. The performance is fantastic. Latency, if we pay attention to the internal tests of Mistral, fantastic. Answer three times faster than the goal model. Its performance is comparable to GPT-4o Mini, and also exceeds QWEN-2.5 32B latency, which is something better in some benchmarks. This model, yes, has just had successor with QWen2.5-Max. Open Source, and European data. Another advantages of Mistral’s AI is that it is an Open Source AI –“Open Weights”, rather-, as he calls, Deepseek or Qwen. Unlike them, Mistral is a startup that raises data governance in the EU, something that can be inert for government agencies and European companies. It has already been seen how there is suspicion about where the data ends From our chats with Deepseek –Italy is investigating In this regard – and this is an undoubtedly striking option. Interesting uses. Mistral developers explain that their model is perfect for conversational assistants, because in them precisely matters that they respond quickly. They also highlight the ability to customize/polish the model to specialize in certain tasks such as legal council, medical diagnosis or technical support. The model is available on platforms such as Hugging Facebut it can also be executed at home in quantized seeing if you have at least 4090 or for example a MacBook with 32 GB of memory. At the moment it does not seem available In le chatthe web service that has confessed to us to be based on Mistral Large 2.1. Reasoning model in sight. But while their competitors are launched to the race for the reasoning models, Mistral has left something behind in this area. In a message in x They clarify that Small 3 does not use synthetic data “which makes it a great base for anything in reason.” In their official announcement they go further and point out that “among many other things, in the coming weeks, large and small Mistral models are expected with greater reasoning capacity.” Deepseek advancing on the right. The Chinese fashion startup has become the star of the moment with its models, and especially with its reasoning model, Deepseek R1. He has also achieved it using the same weapon as Mistral: efficiency. Deepseek’s success validates Mistral’s strategy, of course, but the question is whether Mistral will continue to compete or be overwhelmed by Chinese and US startups if they also partially or totally adopt that same approach. Mistral’s market share is modest, but is in danger of reducing with the strong competition of US and China companies. Source: FT. Doubts about the future. In media as Financial Times The concern about the future of Mistral and the European startups is discussed. There is no other to work in LLM –The German Aleph Alpha He left that race In September 2024-, and that compromises the future of European efforts to compete and not depend completely on the models of China and the US no matter how they can be. In Spain projects such as Alia are an interesting first step, but for now they are far from the LLMs and models mentioned. Operational limbo. This economic newspaper also indicates that with an assessment of 6,000 million euros Mistral is in a kind of limbo. He has raised too much money to gradually disappear, but not enough to compete with US mega -companies, for example. Acquihire on the horizon? Sean Maher, from the Entext consultancy, believes that Mistral can Follow the steps of inflection aiwhich was acquired by Microsoft – Mustafa Suleyman, current head of Ia in Redmond, was his co -founder. Thus, a potential acquisition of its resources and especially of their talent (which is usually called ‘acquihire’) is not ruled out. Image | Mistral ai

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.