Dimension scores are derived from public data and fields; weighted into the composite. Reference only.
Merlon.ai is a localized AI assistant service for doctors’ practices, law firms, and tax advisory firms in Germany. It is not a typical SaaS tool; instead, it helps clients deploy open language models on their own in-office hardware, enabling AI to handle communications, case files, to-dos, receipts, and appointment preparation. It emphasizes that data does not leave the local system and is designed around professional confidentiality requirements such as §203 StGB.
The website states that it can run open models such as Llama, Qwen, and Mistral on local devices, including Apple Silicon or NVIDIA GPU machines, without external API calls or cloud inference. Typical capabilities include summarizing long documents, searching across case files, generating letters from templates, drafting emails, pre-checking invoices, and preparing schedules. On the data side, documents are indexed and processed locally, with no uploads to third parties, no use of customer data for model training, no telemetry, and audit logs for every request. This is a core value proposition for users bound by professional confidentiality, such as lawyers, doctors, and tax advisors.
Its commercial model is mainly project-based, covering workflow assessment, use-case identification, compliance checks, hardware and model selection, integration, documentation delivery, and staff training. The main content does not disclose specific pricing for one-off project packages. It only states that optional ongoing operations are available for 850–3,500 euros/month, split into Basis, Business, and Premium tiers by response time, model updates, and quarterly communication. Additional consulting is 180 euros/hour, and development is 140 euros/hour, both plus tax. No free tier or online trial is mentioned.
The main advantage is its very clear positioning: it focuses on highly sensitive data scenarios where cloud AI is difficult to adopt, uses local deployment and open-source models, presents a complete compliance narrative, and can be customized around client templates, writing style, and existing IT. The drawbacks are that the initial deployment threshold is high, requiring hardware, integration, and operations support; project pricing is not fully transparent; and the website does not provide model performance benchmarks, real customer cases, or delivery timelines. Chinese-language support is also not clearly stated. Although Qwen is mentioned, that alone does not confirm Chinese service capabilities.
It is best suited to law firms, tax advisors, and doctors’ practices in Germany, as well as organizations with strict confidentiality requirements that are not suitable for sending documents to cloud-based large models. For Chinese users who simply need a Chinese knowledge base or office assistant, self-hosted Dify, Open WebUI, AnythingLLM, or a local RAG solution based on Qwen may be a better fit. The site does not disclose access or payment availability from mainland China, so this should be considered unknown; its legal compliance focus is also clearly centered on German professional regulations.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on merlon.ai official site.
merlon.ai is an Germany AI Apps provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of Workable. Click "Visit Official Site" to reach merlon.ai directly.