This model fits the mistral-medium family on SAP AI Core, presented as a closed-weights instruction-tuned endpoint that accepts both text and image inputs while returning text outputs. The placement in the mistral-medium lineup suggests a mid-tier balance of capability and cost rather than a flagship or a lightweight option, with the directory sources aligning it alongside other recent medium-tier Mistral releases. Practical applications span multimodal assistants, document and image-aware workflows, and agent-style integrations that call external services, since the endpoint exposes native tool calling, attachment handling, and adjustable temperature for response control.

With the cataloged API limit context window matched by an equally large the cataloged API limit maximum output, the endpoint is suited to long-document analysis, multi-turn reasoning over attached material, and broader conversational tasks without aggressive truncation. The pricing position of $the listed price per million input tokens and $the listed price per million output tokens undercuts several sibling medium-tier deployments listed in the same directory, making it a reasonable pick for production workloads that need vision support and tool use without stepping up to the most expensive tiers. Integration is documented through SAP AI Core docs and a community SDK, lowering the barrier for teams already operating inside the SAP ecosystem who want a multimodal instruct model with reliable function-calling behavior.