Mistral Unveils Advanced AI Model 3.1
On Monday, Mistral, a Paris-based artificial intelligence firm, launched its latest model, Mistral Small 3.1. This new iteration introduces two open-source variants—chat and instruct—boasting enhanced text performance and multimodal capabilities. The company asserts that the model surpasses competitors like Google’s Gemma 3 and OpenAI’s GPT-4o mini in various benchmarks, particularly highlighting its rapid response times.
Mistral Small 3.1 AI Model Released
The Mistral Small 3.1 model was detailed in a recent announcement from the company. It features an expanded context window of up to 128,000 tokens and delivers impressive inference speeds of 150 tokens per second. This rapid response capability makes it suitable for a range of applications. The model is available in two distinct variants: the chat version functions as a standard chatbot, while the instruct variant is specifically fine-tuned to adhere to user commands, making it ideal for application development with targeted functionalities. As with previous models, Mistral Small 3.1 is accessible to the public. Users can download the open weights from the firm’s Hugging Face listing. The model is distributed under an Apache 2.0 license, permitting academic and research use but prohibiting commercial applications. This accessibility allows a broader audience, including those without high-end computing setups, to experiment with the model.
Optimized for Accessibility and Performance
Mistral has designed the large language model (LLM) to operate efficiently on a single Nvidia RTX 4090 GPU or a Mac device equipped with 32GB of RAM. This optimization ensures that enthusiasts and developers can easily access and utilize the model without needing expensive hardware. Additionally, the model supports low-latency function calling and execution, which can enhance automation and agentic workflows. Developers are also encouraged to fine-tune the Mistral Small 3.1 for specialized domain applications, further expanding its usability.
Benchmark Performance Highlights
In terms of performance, Mistral has shared various benchmark results from internal testing. The Mistral Small 3.1 reportedly outperforms both Gemma 3 and GPT-4o mini across several benchmarks, including the Graduate-Level Google-Proof Q&A (GPQA) Main and Diamond, HumanEval, MathVista, and DocVQA. However, it is worth noting that GPT-4o mini excelled in the Massive Multitask Language Understanding (MMLU) benchmark, while Gemma 3 showed superior performance in the MATH benchmark. Beyond Hugging Face, the new model will also be accessible through the application programming interface (API) on Mistral AI’s developer platform, La Plateforme. Additionally, it will be available on Google Cloud’s Vertex AI, with plans to launch on Nvidia’s NIM and Microsoft’s Azure AI Foundry in the coming weeks.
Observer Voice is the one stop site for National, International news, Sports, Editor’s Choice, Art/culture contents, Quotes and much more. We also cover historical contents. Historical contents includes World History, Indian History, and what happened today. The website also covers Entertainment across the India and World.
Follow Us on Twitter, Instagram, Facebook, & LinkedIn
