Mistral AI Releases Mistral Large and Closes $415M Series A

Summary

Mistral AI released Mistral Large, its first proprietary (non-open-weight) model, alongside the announcement of a $415 million Series A round at a $2 billion valuation. The shift from fully open to partially proprietary models marked a strategic pivot that surprised and disappointed some open-source advocates.

What Happened

On February 26, 2024, Mistral AI released Mistral Large through its new commercial platform, La Plateforme, and via an exclusive partnership with Microsoft Azure. Unlike Mistral 7B and its Mixtral models, which had been released under permissive open licenses, Mistral Large was a proprietary model available only through API access.

Mistral Large was positioned as competitive with GPT-4, with particularly strong multilingual capabilities reflecting the company's European focus. The model featured a 32K token context window and was fluent in English, French, Spanish, German, and Italian, among other languages.

The same week, Mistral announced that it had closed a $415 million Series A funding round, valuing the company at approximately $2 billion — less than a year after its founding. Investors included Andreessen Horowitz, LightSpeed Venture Partners, and existing backers.

Why It Matters

Mistral's pivot to proprietary models illustrated a recurring tension in AI business strategy. The company had built its reputation and community goodwill through aggressive open-source releases, but its investors expected returns that were difficult to generate from freely available models alone. The release of Mistral Large as a closed model was a pragmatic business decision but felt to many in the open-source community like a betrayal of the company's founding ethos.

The Microsoft partnership — providing exclusive cloud distribution for Mistral's proprietary models — echoed the hyperscaler-AI lab partnership pattern established by Microsoft-OpenAI and Amazon-Anthropic. It suggested that even companies founded on open-source principles found it difficult to resist the gravitational pull of cloud distribution partnerships.

Mistral's trajectory from open-source champion to hybrid open/proprietary company became a case study in the economics of open-weight AI: the models may be free, but the companies that build them are not.