Mistral AI has introduced a new free tier aimed at developers, allowing them to fine-tune and create test applications using the startup’s AI models. The announcement, made in a blog post on Tuesday, also revealed significant price reductions for accessing its models through API endpoints. Additionally, Mistral's consumer chatbot, le Chat, now features image processing capabilities.
The move aligns with a growing trend among AI providers—offering more tools at lower prices. Recently valued at $6 billion, the Paris-based company is working to attract developers by making its advanced large language models (LLMs) more accessible. As LLMs become increasingly commoditized, Mistral faces stiff competition from players like OpenAI, Google, and Anthropic, all of which have also introduced free tiers with limited rate access.
Mistral’s free tier, available through its API platform la Plateforme, lets developers experiment and prototype at no cost. However, those seeking to use Mistral’s models in production environments may need to upgrade to a paid commercial plan with higher rate limits.
Although Mistral provides its AI models openly under the Apache 2.0 license, the complexities of hosting them have led the company to offer this free tier—easing the path for developers while positioning its platform as a gateway to more advanced usage.
Further enticing developers, Mistral has slashed prices for its models. Access to Mistral NeMo, Mistral Small, and Codestral via API endpoints is now more than 50% cheaper, with a 33% reduction for Mistral Large.
The startup also expanded its multimodal capabilities with the introduction of Pixtral 12B, its first model to process both images and text. Now available on le Chat, this model allows users to analyze, scan, and search image files, broadening the functionality of Mistral’s tools.
