We’re releasing the weights of Mistral Large 4
Mistral released Mistral Large 4 (Le Chonk), a 1-trillion-parameter multimodal model, now available with improved coding, cybersecurity, and reasoning efficiency. It leads open-weight benchmarks in coding and defensive security while offering three reasoning modes for cost-performance balance. Fully open-weight and European-trained, it empowers sovereign AI with customizable, self-hosted deployment options.
We release the weights of Le Chonk, advancing the frontier of open models in agentic coding and cybersecurity.
Earlier this month, we released Mistral Large 4 (Le Chonk), our flagship 1-trillion-parameter, natively multimodal model, in public preview. The response to the preview has exceeded our expectations, and we’re grateful to our developer community, early-access customers, and partners for their support, collaboration, and feedback.
Since then, we’ve put that feedback to work while continuing to scale reinforcement learning, driving massive improvements in coding and defensive cybersecurity. Gains continue to accelerate, with no signs of saturation.
Today, we’re releasing the weights of this improved Mistral Large 4. The new model also features substantial gains in reasoning efficiency that give developers greater performance while lowering inference costs.
Mistral Large 4 ranks as the best open-weight model for coding and among the world’s most capable open models overall. Trained from scratch on European infrastructure with clear lineage from inception to release, it gives organizations a trusted foundation for Sovereign AI
Mistral is defining the sovereign AI category and leading the open frontier. We’re committed to building the world’s most capable models and releasing them openly, giving developers, enterprises, and institutions the power to own, customize, and control their AI.
Mistral Large 4 entered public preview with strong results across leading coding benchmarks. Just three weeks of additional reinforcement learning have since propelled it to the forefront of open-weight coding, taking the lead on FrontierCode and Terminal-Bench while matching the best open-weight models on the Artificial Analysis Coding Agent Index with substantially greater token efficiency.
On Cognition’s private FrontierCode 2.0 benchmark, which assesses end-to-end code quality and whether a repository maintainer would merge the agent’s change, Mistral Large 4 leads with a score of 19.7%.
Mistral Large 4 delivers frontier coding performance with significantly fewer tokens, making sophisticated coding agents more economical to run at scale.
On the Artificial Analysis Coding Agent Index, Mistral Large 4 scores 48%, effectively tied with the leading open-weight model, GLM-5.3, while using 47% fewer tokens per task on average.
Coding Agent Index scores and token efficiencies for reasoning levels available on Artificial Analysis
The gains extend to terminal-based coding workflows. On Terminal-Bench 4.0, Mistral Large 4 improved by 13 percentage points to 41% compared to its public preview, taking the lead among open-weight models while reducing output tokens by 30%.
To give developers greater control over this tradeoff between performance and cost, Mistral Large 4 now supports three reasoning modes:
Simple tasks; replaces ‘None’ reasoning mode
Hardest use cases requiring high precision and thorough reasoning
Beyond benchmarks, Mistral Large 4’s coding capabilities are validated by human preference. In Surge Code’s blind pairwise evaluations, Mistral Large 4 was preferred over leading open-weight models including GLM-5.3, Kimi K3, and DeepSeek V4.1 Flash, ranking behind only Claude Opus 5.5 among the models evaluated.
Blind side-by-side win rates against other models
Share of prompts where graders preferred Mistral Large 4’s answer over the opponent’s.
Since its preview, we’ve been working with leading cybersecurity vendors to evaluate Mistral Large 4 in real-world environments to test Mistral Large 4’s strengths in vulnerability reproduction and patching.
Mistral Large 4 scores 51% on the Artificial Analysis Cyber Index, which measures a model’s ability to discover vulnerabilities in a codebase, reproduce and validate them, and patch them without breaking existing functionality. Crucially, Mistral Large 4 handles these legitimate defensive security tasks without unnecessary refusals.
Safety has improved alongside capability. Our model’s built-in guardrails distinguishes between defensive and malicious, offensive cybersecurity tasks. It retains strong performance on Cyber benchmarks, while achieving an average refusal rate of 96.6% on malicious cyber requests, the highest we measure among open models.
Average refusal on the cyber-related prompts of JailbreakBench, StrongREJECT and AgentHarm
Here’s what global leaders in cybersecurity have to say about Mistral Large 4:
Testing by Booz Allen demonstrated the success of the European Mistral Large 4 model against our internal benchmark for source code security analysis. The model was able to match vulnerability detection recall rates against other frontier cyber models with a compelling cost curve. And, Mistral’s support for model customization and data-sovereign deployment also lets us tailor it for our mission needs and client environments with strict data-residency requirements. We look forward to working with Mistral to strengthen critical infrastructure around the globe.
Brad Medairy, President National Cyber at Booz Allen Hamilton
Organizations that treat transparency and openness as strengths are positioned well as AI becomes central to security operations. Open-weight cybersecurity models extend that transparency to enterprises and governments that need to host and customize AI under their own control, and they allow researchers to study how these models behave, including where they could be misused. As adversaries increasingly use AI to accelerate their operations, we aim to work alongside model developers such as Mistral to help security teams build on that control, establishing a foundation for resilience by design rather than reactive threat management.
Amy Chang, Head of Cisco’s AI Threat Intelligence and Security Research
When threats move at machine speed, human defense cannot keep pace. Worse, a defense you don’t control isn’t defense—it’s exposure. Security teams perform best when they take the steering wheel. Mistral’s ML4 lets defenders audit the architecture, deploy on their own infrastructure, and keep full sovereignty. Resilience starts with autonomy, and open-weight models help give organizations the control to lead their own defense.”
Stephanie Cohen, Chief Strategy Officer at Cloudflare
Powerful AI models alone won’t tip the scales in cybersecurity. It takes deep security expertise, real-world threat intelligence, and rigorous testing. Our work with Mistral Large 4 tests how well frontier AI can reason through complex vulnerabilities, accelerate remediation, and ensure defenders move faster than adversaries.
Dr. Bartley Richardson, Chief AI and Autonomous Systems Officer, CrowdStrike
Elastic gives security teams the freedom to bring the AI they trust to the data they control, including Mistral’s open models, which are supported on Elastic Security today. We’ve already evaluated ML4 against real cybersecurity workloads, where models need to reason across complex data to help analysts make informed investigation and response decisions. This demonstrates the partnerships we are building to deliver powerful, open-weight models combined with the security context, transparency and control our customers demand.
James Spiteri, Senior Director, Product Management, Elastic
ENISA is experimenting with closed and open weight models in order to better understand their cyber capabilities. Open-weight models allow organisations to keep control of the technology their resilience depends on. ENISA access to and are testing Mistral Large 4 to identify security vulnerabilities, to audit code and analyse defensive logs.
Laura Heuvinck, Head of Communications of ENISA
We believe open-weight AI models have an important role to play in advancing cybersecurity by giving organizations greater flexibility, choice, and control. Innovations like Mistral Large 4 represent promising steps forward in AI’s potential to transform cyber defense. We see continued AI innovation as an important opportunity to strengthen cybersecurity and help organizations navigate an evolving threat landscape.
John Whittle, Chief Operating Officer at Fortinet
Open models let security researchers and developers inspect, adapt and build on AI. Mistral is expanding what the ecosystem can build, and our collaboration on NVIDIA accelerated infrastructure helps developers put these models to work.
Kari Briski, VP of Open Models and Software, NVIDIA
As threats become ever faster, industrialized and now automated, defense must operate at a different scale. Volume, speed, trust: these are the three challenges facing cybersecurity in the age of AI. With Mistral AI, we are building European cyber AI, trained by our experts and operated under their control. By combining the power of AI, the expertise of our teams and our Cyber Threat Intelligence, we are accelerating our response so that we can protect more broadly, more quickly and with trusted European technology.
Hugues Foulon, CEO of Orange Cyberdefense
Rubrik supports Mistral and the push for ‘sovereign AI,’ to give European governments and regulated industries a credible option that does not route their data to other countries. Customers want model choices, including open-weight and private models, to support data residency, latency, cost and capability requirements. As part of our focus to deliver cyber resilience and speed recovery from both AI-powered attacks outside an organization and internal agents, we’ll continue rigorous cybersecurity testing with Mistral Large 4 and work closely together to serve evolving regional demands.
Anneka Gupta, Chief Product Officer of Rubrik.
As enterprises bring AI into production, they increasingly want models they can run on their own terms, with clear visibility into model behavior, data access, and agent permissions. Mistral Large 4 brings strong cybersecurity and technical reasoning to open weight AI, and many European enterprises are already embracing it as their sovereign model of choice. Zscaler Security for AI offerings, delivered on our sovereign Zero Trust Exchange in Europe, now integrate with Mistral, helping organizations continuously test their AI deployments, map what every model and agent can reach, and enforce policy inline. Together, we are empowering enterprises to adopt sovereign AI with the visibility, control, and data protection they need to innovate securely.
Dhawal Sharma, EVP of AI and Data Security, Zscaler
In the time since the preview, we carefully monitored cybersecurity-related use in our live traffic. Though we observed the model refusing to comply with black-box attacks against live targets, we see a potential for uplift to conduct real attacks. We are modifying our guardrails accordingly, and will continue updating them in the months to come.
We are also providing enterprise design partners such as Orange Cyberdefense, with preview access to a specialized variant of Mistral Large 4 optimized for advanced cybersecurity workflows, including penetration testing, vulnerability assessment, and red-teaming. Access is subject to rigorous vetting and contractual safeguards, with testing restricted to owned or explicitly authorized systems in controlled environments.
Meanwhile, customers are tailoring Mistral Large 4 with proprietary security data, policies, detection logic, and threat environments using Mistral Forge. For well-defined cybersecurity tasks, it can be further optimized into smaller, task-specific variants for greater accuracy, making security applications more effective.
Learn more about sovereign AI for cybersecurity and how to work with us.
Mistral Large 4 has continued improving on a range of broad enterprise-use metrics.
For finance-related work, we improved the model’s ability to perform real-world workflows, especially ones involving the creation and editing of spreadsheets for finance and accounting use cases. This is reflected in our Finch scores climbing to 73.3%, second only to Kimi K3 for open models.
Web browsing is a critical part of a majority of knowledge work. Our RL pipeline enabled us to reach 84.5% on BrowseComp (starting from an SFT checkpoint that was achieving 75%). This brings Mistral Large 4 at a higher performance than models like GLM-5.3 or DeepSeek Flash 4.1 1017.
Agents deployed by an organization need to be secure. Whether deployed to internal members or external users, agents shouldn’t open you up to threats like system prompt exfiltration or malicious code injection. RL increased our gains on the B3 Agent Security Benchmark, with Mistral Lar