No connector count stated. Context Chat indexes user files and 'content providers' items'; external storages supported but 'may not work as well'.
Limitations
Language models 'are likely to generate false information'; multi-tenant installation not supported; files >100 MB and password-protected files not supported; embedding model cannot be changed after install; tested only with GPT-3.5 and Llama 3.1 8B ('no guarantees' for others); support covers app bugs not model output
Model testing limitation from https://docs.nextcloud.com/server/stable/admin_manual/ai/app_assistant.html.
Fit and decision conditions
Prerequisites
Nextcloud 32+; AppAPI Deploy Daemon (Docker/Podman); NVIDIA (CUDA 12.8+) or AMD (Vulkan) GPU with >=2 GB VRAM and 8 GB RAM, or x86_64 CPU with AVX/AVX2 and >=12 GB RAM; dedicated machine recommended; a text-processing provider; Ultimate subscription for support
Quote: 'As a customer, you can gain access to the Nextcloud AI Assistant through our Ultimate subscription.' (https://nextcloud.com/assistant/). Talk plan lists Assistant as 'Optional (extra costs)'.
Published price
Ultimate from EUR 204.75/user/year at 100 users (EUR 183.75 at 200 users); Standard EUR 71.29 and Premium EUR 104.99 exclude the Assistant
Net, excl. VAT. Open-source community edition exists; price applies to Enterprise subscription.
Data and controls
Where data is stored
Customer-determined: runs on customer's own server ('Works locally on your server'); optional European LLM-as-a-service partners Deutsche Telekom, IONOS, OVHcloud, plusserver
Quote (on-prem): 'data stays local and is not leaked to third parties through training of new AI models.' Applies to local models; training policy of optional OpenAI/Gemini/partner backends not detailed. Admin manual makes no training statement.
Certificates and audit reports
None claimed on security page; 'aligns with' ISO/IEC 27001-2013 Clause 14 (alignment, not certification); code audit by Kyos for city of Geneva; NCC Group security review
No certification with issuer found on assistant, pricing or security pages.
Additional data conditions
Subprocessors and underlying models
Self-hosted: no Nextcloud sub-processor in data path for local models. Models: Nextcloud-built and open-source local LLMs (llm2); optional OpenAI, Dall-E, Gemini, IBM watsonx, Replicate; LLM-as-a-service from Deutsche Telekom, IONOS, OVHcloud, plusserver. Bring-your-own (local) model supported via llm2/LocalAI
Company/support: 'Our support is in English. Other languages possible through partners'; Assistant languages depend on configured text model; German not explicitly mentioned
Language support depends on the configured text-to-text provider.
Existing permissions
Partial: Context Chat syncs share/ownership changes, but quote 'Context Chat does not follow these rules' (files_accesscontrol rules); users denied by access-control may still reach documents visible to them in Files
No explicit statement that query-time results are filtered by per-user file access; product page only says scope can be limited to a folder or file.
Questions for the provider
Ask the provider which retention and deletion terms apply, including when the contract ends.
Cost per accepted outcome
Editable hypothetical example · not provider pricing or measured performance. Use this monthly worksheet to test your own quote and quality assumptions.
/ accepted outcome
accepted outcomes / month
5,000 tasks × 80% = 4,000 accepted outcomes per month. Higher entered costs raise unit cost; more accepted outcomes spread the same cost over more results.
Formula: entered monthly costs ÷ (monthly tasks × acceptance % ÷ 100). The numerator includes only the costs you enter. Add applicable licence, usage, integration, human review and rework costs yourself; this is not automatically a fully loaded estimate or a vendor billing model. Capacity released is not realized financial value.
Inputs stay in this browser tab’s session, including reload and language changes. They are not added to the page URL. Continuing prepares an unsent enquiry draft.
A written comparison of up to three AI offerings · €195 net
You get the facts that matter for your decision, each with its source, the questions to ask each vendor and a cost estimate based on your own numbers. First you agree your three most important criteria with Sten Franke.
We deliver your brief by email within a maximum of five working days after payment is confirmed and we have agreed the question and minimum information needed to begin. ETHORITY confirms the start date and delivery date and delivers earlier when the reviewed brief is ready. Missing optional information is recorded and assessed; it does not require you to research the solution yourself. Review before delivery and email support are included.
ETHORITY®