Home · Technology · Oct 7 archive
French AI Firm Mistral Launches Le Chonk to Challenge US and Chinese Models
Confirmed
In Short: Mistral's chief scientist, Guillaume Lample, says Le Chonk can challenge the best models despite the company's limited resources compared to competitors like OpenAI and Anthropic.
French company Mistral has released a new open-source artificial intelligence model called Le Chonk, which it claims can compete with the best proprietary models from the US and China.
Le Chonk, officially named Mistral Large 4, has 1.05 trillion parameters and is optimized for tasks such as coding, cybersecurity, and specific industrial applications.
The model is designed to be open-weight, meaning it can be freely used and customized by anyone, potentially reducing costs for businesses.
The model's open-source nature and performance claims come at a time when the US government has placed temporary restrictions on the distribution of models from OpenAI and Anthropic due to cybersecurity concerns.
Le Chonk is optimized for cybersecurity tasks, including log inspection, code auditing, and identifying newly published exploits or vulnerabilities.
Mistral says the model is particularly strong in agentic capabilities, which are crucial for handling complex tasks like coding and analyzing requests.
The company trained Le Chonk from scratch over two months using 4,000 Nvidia Grace Blackwell GPUs hosted in European data centers.
Le Chonk is available in a public beta through Mistral’s API in Mistral Studio, with the weights to be published in an open-weight format on October 27.
Andrea Renda, director of research at the Centre for European Policy Studies, says Mistral's position is favorable due to the EU's push for technological sovereignty and the US's increased hostility towards AI models.
Mistral aims to eliminate the few remaining reasons businesses might hesitate to choose open-source models by reducing the performance gap with proprietary models and providing a competitive alternative to Chinese models.
What this adds
The nickname 'Le Chonk' for the model is consistent across multiple outlets.
The model's release comes amid heightened tensions over AI model access and distribution.
What's confirmed
- French company Mistral has released a new open-source artificial intelligence model called Le Chonk, which it claims can compete with the best proprietary models from the US and China.
- Le Chonk, officially named Mistral Large 4, has 1.05 trillion parameters and is optimized for tasks such as coding, cybersecurity, and specific industrial applications.
- The model is designed to be open-weight, meaning it can be freely used and customized by anyone, potentially reducing costs for businesses.
- The model's open-source nature and performance claims come at a time when the US government has placed temporary restrictions on the distribution of models from OpenAI and Anthropic due to cybersecurity concerns.
- Le Chonk is optimized for cybersecurity tasks, including log inspection, code auditing, and identifying newly published exploits or vulnerabilities.
- Mistral says the model is particularly strong in agentic capabilities, which are crucial for handling complex tasks like coding and analyzing requests.
- The company trained Le Chonk from scratch over two months using 4,000 Nvidia Grace Blackwell GPUs hosted in European data centers.
- Le Chonk is available in a public beta through Mistral’s API in Mistral Studio, with the weights to be published in an open-weight format on October 27.
- Andrea Renda, director of research at the Centre for European Policy Studies, says Mistral's position is favorable due to the EU's push for technological sovereignty and the US's increased hostility towards AI models.
- Mistral aims to eliminate the few remaining reasons businesses might hesitate to choose open-source models by reducing the performance gap with proprietary models and providing a competitive alternative to Chinese models.
What's still developing
- “There are a lot of areas where the other labs will not focus that much,” Guillaume Lample, cofounder and chief scientist at Mistral, tells WIRED.
- Though Le Chonk is built to compete with leading general-purpose models, it’s optimized specifically for coding and cyberdefense, as well as tasks particular to manufacturing, finance, electrical engineering, and other niches.
- “There are so many domains in which you can improve models.” Mistral presents Le Chonk as by far the most capable open-weight model developed outside of China, and “very, very close” to some proprietary models.
- It’s already considerably cheaper for businesses to run open-weight models, which cost only as much as the compute they consume.
- “Mistral is still in the race of getting the best model,” Lample says.
- “This is the main message.” With less capital and fewer compute resources than OpenAI and Anthropic, Mistral has generally lagged behind on model performance, revenue, and the frequency of releases.
- New AI models are coming, and they don’t stop coming.
- On the heels of the recent rollout of Claude Opus 5.5 and OpenAI’s GPT-6 Sol and GPT-6 Luna, Mistral AI is launching a new flagship model, Mistral Large 4, with a public preview available on Tuesday.
- A French company known for its open-source and open-weight LLMs, Mistral AI reports that the Large 4 model “ranks among the best open-weight models of the world” on aggregated metrics and also outperforms European and US models, while also catching up to Chinese competitors, who have been dominating the open-weight sector.
- Lample and Stock said that Mistral Large 4 was built to process both text and images, closing performance gaps from previous Mistral models.
- “Really, the core competency we’re looking at for ML4 is agentic capabilities,” said Stock.
- The Mistral AI team said the model reaches “state-of-the-art” performance on critical agentic benchmarks like SWE-bench and Terminal Bench – evaluation frameworks for testing how well agents handle complex tasks – for tasks like coding, analyzing and pulling requests.