There's Mistral. Although they've been surprisingly quiet for the last 6 months.
Mistral: eq.IQ ~70
Kimi, Claude, Grok, Gpt, Gemini: eq.IQ ~130+
Open weights is more beneficial to Europe spring-boarding a catch up than defending Claude/Anthropic.
They seem to be oriented more toward customizing models for the concrete needs of a company, on-prem deployment, proprietary knowledge-bases, etc.
I agree, but you are underestimating EU regulatory incompetence