Mistral Releases an Open Source Model that Outperforms Gemma 3 and GPT-4o Mini
- Mistral AI has released Mistral Small 3.1, an open-source multimodal model that outperforms comparable models like Gemma 3 and GPT-4o Mini.
- The model has an expanded context window of up to 128k tokens and a delivery inference speed of 150 tokens per second, making it competitive with larger models like Llama 3.3 70B or Qwen 32B.
- Mistral Small 3 can be fine-tuned to specialize in specific domains, creating highly accurate experts, particularly useful in fields like legal advice, medical diagnostics, and technical support.
- The release of Mistral Small 3.1 marks a significant milestone for the company, which has raised $1.04 billion and established itself as Europe’s leading AI startup with a valuation of approximately $6 billion.
- Mistral AI’s open-source approach highlights a growing divide in the AI industry between closed, proprietary systems and open, accessible alternatives, setting the stage for increased competition in the market dominated by U.S. tech giants.
IBL News | New York
Parisābased Mistral AI unveiledĀ Mistral Small 3.1, a new multimodal open-source model. According to the company, it isĀ “the best model in its weight class ”Ā and “outperforms comparable models like Gemma 3 and GPT-4o Mini. “
Released under an Apache 2.0 license, Mistral Small 3.1 has an expanded context window of up to 128k tokens and a delivery inference speed of 150 tokens per second.
Experts say that Mistral Small 3 is competitive with larger models such as Llama 3.3 70B or Qwen 32B and replaces opaque proprietary models like GPT4o-mini.
Mistral Small 3 can be fine-tuned to specialize in specific domains, creating highly accurate experts. This is particularly useful in fields like legal advice, medical diagnostics, and technical support, where domain-specific knowledge is essential.
This model sets theĀ stage for increased competition in a market dominated by U.S. tech giants. Mistral’s open-source approach highlights a growing divide in the AI industry between closed, proprietary systems and open, accessible alternatives.
After raising $1.04 billion, founded in 2023 by former researchers from Google DeepMind and Meta, Mistral AI has rapidly established itself as Europeās leading AI startup, with a valuation of approximately $6 billion. While impressive for a European startup, this valuation remains a fraction of OpenAIās reported $80 billion.
Mistral Small 3.1 joins the company’s rapidly expanding suite of AI products.
Earlier this month, the company introducedĀ Mistral OCR, an optical character recognition API that converts PDF documents into AI-ready Markdown files. This addresses a critical need for enterprises seeking to make document repositories accessible to AI systems.
These specialized tools complement Mistralās broader portfolio, which includesĀ Mistral Large 2Ā (their flagship large language model),Ā PixtralĀ (for multimodal applications),Ā CodestralĀ (for code generation), and āLes Ministraux,ā a family of models optimized for edge devices.