Mistral AI Opens Vibe Platform to Third-Party Open Models, Starting With Z.ai GLM-5.2
The move lets Vibe users pick models beyond Mistral's own lineup, broadening the platform's appeal as a neutral host for open-weight AI.
Reporting from 1 source: GIGAZINE.
Mistral AI announced it will support third-party open models on its platform, including the Vibe integrated environment, starting with Z.ai's GLM-5.2. The company also launched general availability of Mistral Regional Endpoints, letting users choose US or European regions for inference, and detailed its Mistral Priority Tier for mission-critical workloads.
Mistral AI is no longer limiting its platform to its own models. The French company said it will support third-party open models in products like Vibe, its integrated environment for chat, coding, and work assistance. The first supported model is Z.ai's GLM-5.2, and cases of running it in Vibe have already been reported.
Alongside the model expansion, Mistral started general availability of Regional Endpoints, which lets users choose whether inference runs in the US or Europe based on data storage, regulatory, and latency needs. The company also noted that its Priority Tier, in public preview, offers both processing region selection and uptime guarantees for mission-critical workloads, which it says is unique among AI labs in Europe.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.