Mistral’s New Model Is Trained to Solve ‘Real GitHub Issues’

Mistral, the French AI startup, released a new open-source model called Devstral on Wednesday. The model was built in collaboration with All Hands AI, a startup building open-source software development agents. 

Devstral is trained to excel at coding-related tasks. On the SWE-Bench Verified benchmark, which evaluates AI models on real-world software issues, Devstral scored 46.8%, outperforming other open-source AI models. It outperforms OpenAI’s GPT-4.1 Mini and Claude 3.5 Haiku. 

Mistral said that the model is trained to solve real GitHub issues and overcomes the challenges of typical large language models, which are excellent at code completion or atomic coding tasks, but struggle to solve real-world engineering problems. 

Source: Mistral 

“Devstral is light enough to run on a single RTX 4090 or a Mac with 32GB RAM, making it an ideal choice for local deployment and on-device use,” added Mistral. The model is available for download on HuggingFace, Ollama, Kaggle, Unsloth, LM Studio starting today. 

Mistral recently announced a strategic partnership with G42, an Abu Dhabi-based technology group. The partnership will focus on co-developing next-generation AI platforms and infrastructure. 

The partnership encompasses the AI value chain, ranging from training AI models and developing AI agents and infrastructure to creating industry-specific applications throughout Europe, the Middle East, and the Global South.

Besides, Mistral will also explore collaborations with the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), the world's first AI university based in Abu Dhabi, across research and development in foundation models, talent development, and AI research. 

A few weeks ago, the company unveiled Mistral Medium 3, which focuses on cost-effectiveness while outperforming competing models like Meta’s Llama 4 Maverick in benchmark tests. 

The company stated that it is designed for enterprise deployment and excels in coding, STEM, and multimodal tasks. It achieves over 90% of Claude Sonnet 3.7’s benchmark scores at significantly lower pricing—$0.40 per million tokens for input and $2 for output.