IONOS AI Model Hub vs. Ollama: data protection per plan

IONOS AI Model Hub (1 plan) and Ollama (5 plans) side by side: DPA, training on inputs, hosting and data retention per plan, based on the providers’ public documents. Without rating.

Last updated: 18 Oct 2026 (latest review date of the tools shown)

Choose another tool Two tools from the same category open their dedicated comparison page.

Tool 1Tool 2Tool 3 (optional)

At a glance

Each value is listed with the plans it applies to. Pick a tier to see only plans of that tier; if plans are chosen in the plan comparison below, the table shows exactly those.

Plan tier

All plans: each value is listed with the plans it applies to.

Overview: each value with the plans it applies to
Feature	IONOS AI Model Hub	Ollama
Provider based in	IONOS Cloud GmbH, Elgendorfer Straße 57, 56410 Montabaur, Germany	Ollama Inc.
Plans recorded	1	5
Hosting	EU option: AI Model Hub	mixed: all 5 plans
Training on inputs	no: AI Model Hub	no: all 5 plans
DPA	yes: AI Model Hub	not documented: all 5 plans
As of	18 Oct 2026	18 Oct 2026

Plans compared

Plan vs. planAll plans

IONOS AI Model Hub planOllama plan

Plans of the same tier are preselected: the tier chosen in the tier switch, otherwise the tier of the plan you pick yourself. Plans you pick yourself stay. The address holds your selection and can be shared.

Comparison of IONOS AI Model Hub AI Model Hub vs. Ollama Team, one plan each, without rating
Feature	IONOS AI Model HubAI Model Hub (usage-based)Tier: API/Developer · IONOS Cloud GmbH · As of 18 Oct 2026	OllamaTeam (Early Access)Tier: Team/Business · Ollama Inc. · As of 18 Oct 2026

differs by plan	yes	not publicly documented
Details: DPA available	yes (data protection package of IONOS Cloud GmbH as DPA under Art. 28 GDPR, version 7.4, as of 06/2023; according to the documentation part of the contract without separate signature and without a product-specific annex for the AI Model Hub) Provider’s DPA	not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)

	no	no
Details: Training on inputs	Terms of use for AI services section 2.3: input and output data only for providing the service under the contract, technical security and troubleshooting (improvement of the services only for SaaS); any further use, 'in particular for training, retraining or otherwise developing AI models, is expressly prohibited' (translated). Documentation 'Data Handling': prompts and outputs are 'not used to train, fine-tune or otherwise improve any model' and are not disclosed to the model developers.	Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.

differs by plan	EU option	mixed
Details: Data residency	Documentation 'Data Handling': 'All inference is performed in IONOS CLOUD data centers in Germany'; access to operational data only by personnel in the EU, no access from outside the EU for operation, maintenance, support or fault diagnosis. Product page: '100% GDPR-compliant and securely hosted in Germany' (translated; provider statement). The model weights come from third-party developers who, according to the documentation, have no access to customer data.	Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.

	list available	list available
Details: List published	IONOS SE (Germany): technical support, support and operation of the AI Model Hub and of data centres (sub-contractor list, as of 10/2026) Documentation 'Data Handling': 'No third-party sub-processors are engaged for AI Model Hub'; model developers only supply the model weights and do not process personal data on the customer's behalf Provider’s list	NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented

differs by plan	Documentation 'Data Handling': the…	Privacy Policy
Details: Data retention	Documentation 'Data Handling': the content of prompts and outputs is not logged and not stored persistently, only held in host and GPU memory during processing; the API is stateless. Metadata per call (timestamp, model, token count) is recorded for billing and operation; operational logs are retained 'in accordance with IONOS CLOUD's internal retention policy' (period not quantified). Data protection package section 5.7: after processing ends, deletion or return at the customer's choice.	Privacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.

differs by plan	per token	from 500 USD/mo
Details: Price	Usage-based, net prices excl. VAT, offer for businesses only (price list IONOS Cloud GmbH, as of 2026-10-18), per 1 million input/output tokens e.g.: Qwen3.5-9B EUR 0.10/0.15; Llama 3.1 8B Instruct EUR 0.15/0.15; Mistral Nemo Instruct EUR 0.15/0.15; Mistral Small 24B Instruct EUR 0.10/0.30; gpt-oss-120b EUR 0.15/0.65; Llama 3.3 70B Instruct EUR 0.65/0.65; Qwen3.8-27 EUR 0.40/2.70; Qwen3.5-397B-A17B EUR 0.60/3.60; Qwen3-Coder-Next EUR 0.15/0.80; LightOnOCR 2 EUR 0.15/0.30. Images: FLUX.1 [schnell] EUR 0.0288 per image; FLUX.2 [klein] 4B EUR 0.013 per first megapixel. Embeddings from EUR 0.01 per 1 million tokens.	from 500 USD per month (pricing page); usage beyond the included credits is billed from a shared team balance as you go. According to the FAQ, Ollama creates a separate team account on joining. The customer is responsible for taxes (Terms).
	yes	yes
Details: AI features included	yes (inference API for language, code, OCR, image, embedding and reranking models from third-party developers, operated on IONOS infrastructure; OpenAI-compatible API, tool calling, RAG)	yes (Pro plus unlimited users, 1,000 USD of usage credits per month shared across the team, centralised billing and administration, priority support, 10 concurrent requests; shared projects "coming soon" according to the pricing page)

	7 sources	4 sources
Details: Sources with retrieval date	AI Model HubProduct page, retrieved 18 Oct 2026 Price list IONOS CLOUD GmbH (DE), section AI Model HubPricing page, retrieved 18 Oct 2026 AI Model Hub – Data HandlingDocumentation, retrieved 18 Oct 2026 AI Model Hub – European Union AI ActDocumentation, retrieved 18 Oct 2026 Terms of use for AI servicesTerms and conditions, retrieved 18 Oct 2026 Data protection package (data processing), version 7.4, as of 06/2023DPA, retrieved 18 Oct 2026 IONOS Cloud sub-contractor list, as of 10/2026Sub-processor list, retrieved 18 Oct 2026 All details on the tool page	PricingPricing page, retrieved 18 Oct 2026 Terms of Service (Last updated: May 2026)Terms and conditions, retrieved 18 Oct 2026 Privacy Policy (Last updated: March 2026)Privacy notice, retrieved 18 Oct 2026 Docs: CloudDocumentation, retrieved 18 Oct 2026 All details on the tool page

Contract / DPA

differs by plan: IONOS AI Model Hub AI Model Hubyes; Ollama Teamnot publicly documented; IONOS AI Model Hub AI Model Hub
yes (data protection package of IONOS Cloud GmbH as DPA under Art. 28 GDPR, version 7.4, as of 06/2023; according to the documentation part of the contract without separate signature and without a product-specific annex for the AI Model Hub)
Provider’s DPA
Ollama Team
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)

Training

: IONOS AI Model Hub AI Model Hubno; Ollama Teamno; IONOS AI Model Hub AI Model Hub
Terms of use for AI services section 2.3: input and output data only for providing the service under the contract, technical security and troubleshooting (improvement of the services only for SaaS); any further use, 'in particular for training, retraining or otherwise developing AI models, is expressly prohibited' (translated). Documentation 'Data Handling': prompts and outputs are 'not used to train, fine-tune or otherwise improve any model' and are not disclosed to the model developers.
Ollama Team
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.

Hosting

differs by plan: IONOS AI Model Hub AI Model HubEU option; Ollama Teammixed; IONOS AI Model Hub AI Model Hub
Documentation 'Data Handling': 'All inference is performed in IONOS CLOUD data centers in Germany'; access to operational data only by personnel in the EU, no access from outside the EU for operation, maintenance, support or fault diagnosis. Product page: '100% GDPR-compliant and securely hosted in Germany' (translated; provider statement). The model weights come from third-party developers who, according to the documentation, have no access to customer data.
Ollama Team
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.

Sub-processors

: IONOS AI Model Hub AI Model Hublist available; Ollama Teamlist available; IONOS AI Model Hub AI Model Hub
IONOS SE (Germany): technical support, support and operation of the AI Model Hub and of data centres (sub-contractor list, as of 10/2026)
Documentation 'Data Handling': 'No third-party sub-processors are engaged for AI Model Hub'; model developers only supply the model weights and do not process personal data on the customer's behalf
Provider’s list
Ollama Team
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented

Retention

differs by plan: IONOS AI Model Hub AI Model HubDocumentation 'Data Handling': the…; Ollama TeamPrivacy Policy; IONOS AI Model Hub AI Model HubDocumentation 'Data Handling': the content of prompts and outputs is not logged and not stored persistently, only held in host and GPU memory during processing; the API is stateless. Metadata per call (timestamp, model, token count) is recorded for billing and operation; operational logs are retained 'in accordance with IONOS CLOUD's internal retention policy' (period not quantified). Data protection package section 5.7: after processing ends, deletion or return at the customer's choice.
Ollama TeamPrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.

Price

differs by plan: IONOS AI Model Hub AI Model Hubper token; Ollama Teamfrom 500 USD/mo; IONOS AI Model Hub AI Model Hub
Usage-based, net prices excl. VAT, offer for businesses only (price list IONOS Cloud GmbH, as of 2026-10-18), per 1 million input/output tokens e.g.: Qwen3.5-9B EUR 0.10/0.15; Llama 3.1 8B Instruct EUR 0.15/0.15; Mistral Nemo Instruct EUR 0.15/0.15; Mistral Small 24B Instruct EUR 0.10/0.30; gpt-oss-120b EUR 0.15/0.65; Llama 3.3 70B Instruct EUR 0.65/0.65; Qwen3.8-27 EUR 0.40/2.70; Qwen3.5-397B-A17B EUR 0.60/3.60; Qwen3-Coder-Next EUR 0.15/0.80; LightOnOCR 2 EUR 0.15/0.30. Images: FLUX.1 [schnell] EUR 0.0288 per image; FLUX.2 [klein] 4B EUR 0.013 per first megapixel. Embeddings from EUR 0.01 per 1 million tokens.
Ollama Team
from 500 USD per month (pricing page); usage beyond the included credits is billed from a shared team balance as you go. According to the FAQ, Ollama creates a separate team account on joining. The customer is responsible for taxes (Terms).
: IONOS AI Model Hub AI Model Hubyes; Ollama Teamyes; IONOS AI Model Hub AI Model Hub
yes (inference API for language, code, OCR, image, embedding and reranking models from third-party developers, operated on IONOS infrastructure; OpenAI-compatible API, tool calling, RAG)
Ollama Team
yes (Pro plus unlimited users, 1,000 USD of usage credits per month shared across the team, centralised billing and administration, priority support, 10 concurrent requests; shared projects "coming soon" according to the pricing page)

Evidence

: IONOS AI Model Hub AI Model Hub7 sources; Ollama Team4 sources; IONOS AI Model Hub AI Model Hub
AI Model HubProduct page, retrieved 18 Oct 2026
Price list IONOS CLOUD GmbH (DE), section AI Model HubPricing page, retrieved 18 Oct 2026
AI Model Hub – Data HandlingDocumentation, retrieved 18 Oct 2026
AI Model Hub – European Union AI ActDocumentation, retrieved 18 Oct 2026
Terms of use for AI servicesTerms and conditions, retrieved 18 Oct 2026
Data protection package (data processing), version 7.4, as of 06/2023DPA, retrieved 18 Oct 2026
IONOS Cloud sub-contractor list, as of 10/2026Sub-processor list, retrieved 18 Oct 2026
All details on the tool page
Ollama Team
PricingPricing page, retrieved 18 Oct 2026
Terms of Service (Last updated: May 2026)Terms and conditions, retrieved 18 Oct 2026
Privacy Policy (Last updated: March 2026)Privacy notice, retrieved 18 Oct 2026
Docs: CloudDocumentation, retrieved 18 Oct 2026
All details on the tool page

Comparison of IONOS AI Model Hub vs. Ollama by feature, without rating
Feature	IONOS AI Model HubIONOS Cloud GmbH · 1 plan · As of 18 Oct 2026	OllamaOllama Inc. · 5 plans · As of 18 Oct 2026

differs by tool	yesonly plan	not publicly documentedall 5 plans
Details: DPA available	AI Model Hub (usage-based) yes (data protection package of IONOS Cloud GmbH as DPA under Art. 28 GDPR, version 7.4, as of 06/2023; according to the documentation part of the contract without separate signature and without a product-specific annex for the AI Model Hub) Provider’s DPA	Free not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller) Pro not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller) Max not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller) Team (Early Access) not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller) Enterprise not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)

	noonly plan	noall 5 plans
Details: Training on inputs	AI Model Hub (usage-based) Terms of use for AI services section 2.3: input and output data only for providing the service under the contract, technical security and troubleshooting (improvement of the services only for SaaS); any further use, 'in particular for training, retraining or otherwise developing AI models, is expressly prohibited' (translated). Documentation 'Data Handling': prompts and outputs are 'not used to train, fine-tune or otherwise improve any model' and are not disclosed to the model developers.	Free Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses. Pro Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses. Max Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses. Team (Early Access) Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses. Enterprise Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.

differs by tool	EU optiononly plan	mixedall 5 plans
Details: Data residency	AI Model Hub (usage-based) Documentation 'Data Handling': 'All inference is performed in IONOS CLOUD data centers in Germany'; access to operational data only by personnel in the EU, no access from outside the EU for operation, maintenance, support or fault diagnosis. Product page: '100% GDPR-compliant and securely hosted in Germany' (translated; provider statement). The model weights come from third-party developers who, according to the documentation, have no access to customer data.	Free Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device. Pro Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device. Max Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device. Team (Early Access) Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device. Enterprise Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.

	list availableonly plan	list availableall 5 plans
Details: List published	AI Model Hub (usage-based) IONOS SE (Germany): technical support, support and operation of the AI Model Hub and of data centres (sub-contractor list, as of 10/2026) Documentation 'Data Handling': 'No third-party sub-processors are engaged for AI Model Hub'; model developers only supply the model weights and do not process personal data on the customer's behalf Provider’s list	Free NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented Pro NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented Max NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented Team (Early Access) NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented Enterprise NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page) Stripe (payment processing, according to the Terms and Privacy Policy) Cloud infrastructure and model inference providers according to the Privacy Policy, without names Sub-processor list: not publicly documented

differs by tool	Documentation 'Data Handling': the…only plan	Privacy Policyall 5 plans
Details: Data retention	AI Model Hub (usage-based)Documentation 'Data Handling': the content of prompts and outputs is not logged and not stored persistently, only held in host and GPU memory during processing; the API is stateless. Metadata per call (timestamp, model, token count) is recorded for billing and operation; operational logs are retained 'in accordance with IONOS CLOUD's internal retention policy' (period not quantified). Data protection package section 5.7: after processing ends, deletion or return at the customer's choice.	FreePrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented. ProPrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented. MaxPrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented. Team (Early Access)Privacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented. EnterprisePrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.

differs by tool	per tokenonly plan	0 USD… (further prices in the details)Free 20 USD/mo… (further prices in the details)Pro 100 USD/moMax from 500 USD/moTeam (Early Access) customEnterprise
Details: Price	AI Model Hub (usage-based) Usage-based, net prices excl. VAT, offer for businesses only (price list IONOS Cloud GmbH, as of 2026-10-18), per 1 million input/output tokens e.g.: Qwen3.5-9B EUR 0.10/0.15; Llama 3.1 8B Instruct EUR 0.15/0.15; Mistral Nemo Instruct EUR 0.15/0.15; Mistral Small 24B Instruct EUR 0.10/0.30; gpt-oss-120b EUR 0.15/0.65; Llama 3.3 70B Instruct EUR 0.65/0.65; Qwen3.8-27 EUR 0.40/2.70; Qwen3.5-397B-A17B EUR 0.60/3.60; Qwen3-Coder-Next EUR 0.15/0.80; LightOnOCR 2 EUR 0.15/0.30. Images: FLUX.1 [schnell] EUR 0.0288 per image; FLUX.2 [klein] 4B EUR 0.013 per first megapixel. Embeddings from EUR 0.01 per 1 million tokens.	Free 0 USD (pricing page). Additional usage credits at token prices per million tokens, e.g. gpt-oss:120b 0.15 USD input and 0.60 USD output; off-peak prices outside 12:00–18:00 UTC on weekdays and all day at weekends. Purchased credits expire after one year; the customer is responsible for taxes (Terms). Pro 20 USD per month or 200 USD per year (16.67 USD per month billed annually) (pricing page); one account per person. The customer is responsible for taxes (Terms). Max 100 USD per month (pricing page); one account per person. The customer is responsible for taxes (Terms). Team (Early Access) from 500 USD per month (pricing page); usage beyond the included credits is billed from a shared team balance as you go. According to the FAQ, Ollama creates a separate team account on joining. The customer is responsible for taxes (Terms). Enterprise Custom (pricing page: "Custom", volume usage pricing; contact via sales)
	yesonly plan	yesall 5 plans
Details: AI features included	AI Model Hub (usage-based) yes (inference API for language, code, OCR, image, embedding and reranking models from third-party developers, operated on IONOS infrastructure; OpenAI-compatible API, tool calling, RAG)	Free yes (cloud models with starter usage credits and access to starter models; all cloud models after buying usage credits; local models unlimited; 1 concurrent request) Pro yes (Free plus 60 USD of usage credits per month, larger pro models, several models concurrently, 3 concurrent requests; Fast mode "coming soon" according to the pricing page) Max yes (Pro plus 300 USD of usage credits per month, early access to the newest models, 10 concurrent requests) Team (Early Access) yes (Pro plus unlimited users, 1,000 USD of usage credits per month shared across the team, centralised billing and administration, priority support, 10 concurrent requests; shared projects "coming soon" according to the pricing page) Enterprise yes (Team plus model access controls, cost budgets for users and API keys, private Slack channel with dedicated support, custom security questionnaires)

	7 sources	4 sources
Details: Sources with retrieval date	AI Model HubProduct page, retrieved 18 Oct 2026 Price list IONOS CLOUD GmbH (DE), section AI Model HubPricing page, retrieved 18 Oct 2026 AI Model Hub – Data HandlingDocumentation, retrieved 18 Oct 2026 AI Model Hub – European Union AI ActDocumentation, retrieved 18 Oct 2026 Terms of use for AI servicesTerms and conditions, retrieved 18 Oct 2026 Data protection package (data processing), version 7.4, as of 06/2023DPA, retrieved 18 Oct 2026 IONOS Cloud sub-contractor list, as of 10/2026Sub-processor list, retrieved 18 Oct 2026 All details on the tool page	PricingPricing page, retrieved 18 Oct 2026 Terms of Service (Last updated: May 2026)Terms and conditions, retrieved 18 Oct 2026 Privacy Policy (Last updated: March 2026)Privacy notice, retrieved 18 Oct 2026 Docs: CloudDocumentation, retrieved 18 Oct 2026 All details on the tool page

Contract / DPA

differs by tool: IONOS AI Model Hubyesonly plan; Ollamanot publicly documentedall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)
yes (data protection package of IONOS Cloud GmbH as DPA under Art. 28 GDPR, version 7.4, as of 06/2023; according to the documentation part of the contract without separate signature and without a product-specific annex for the AI Model Hub)
Provider’s DPA
Ollama
Free
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)
Pro
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)
Max
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)
Team (Early Access)
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)
Enterprise
not publicly documented (no DPA on the retrieved pages; the Privacy Policy names Ollama Inc. as controller)

Training

: IONOS AI Model Hubnoonly plan; Ollamanoall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)
Terms of use for AI services section 2.3: input and output data only for providing the service under the contract, technical security and troubleshooting (improvement of the services only for SaaS); any further use, 'in particular for training, retraining or otherwise developing AI models, is expressly prohibited' (translated). Documentation 'Data Handling': prompts and outputs are 'not used to train, fine-tune or otherwise improve any model' and are not disclosed to the model developers.
Ollama
Free
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.
Pro
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.
Max
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.
Team (Early Access)
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.
Enterprise
Pricing page (FAQ Privacy): "Prompt or response data is never logged or trained on." Terms (May 2026, section 5): Ollama does not use inputs or outputs to train AI models. Privacy Policy (March 2026): no use of inputs or outputs for training; prompt or response content is not requested in support requests. According to the pricing page, Ollama requires no-logging, no-training and zero-data-retention policies from hosting partners. Local use: according to the Privacy Policy, Ollama has no access to prompts and responses.

Hosting

differs by tool: IONOS AI Model HubEU optiononly plan; Ollamamixedall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)
Documentation 'Data Handling': 'All inference is performed in IONOS CLOUD data centers in Germany'; access to operational data only by personnel in the EU, no access from outside the EU for operation, maintenance, support or fault diagnosis. Product page: '100% GDPR-compliant and securely hosted in Germany' (translated; provider statement). The model weights come from third-party developers who, according to the documentation, have no access to customer data.
Ollama
Free
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.
Pro
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.
Max
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.
Team (Early Access)
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.
Enterprise
Pricing page (FAQ Privacy): Ollama hosts models and compute resources primarily in the United States; to serve global demand, traffic may be routed to Europe ("may route to Europe for additional capacity"). A selectable EU region is not documented. Privacy Policy: data may be transferred to and processed in the United States. Local models run on the user's own device.

Sub-processors

: IONOS AI Model Hublist availableonly plan; Ollamalist availableall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)
IONOS SE (Germany): technical support, support and operation of the AI Model Hub and of data centres (sub-contractor list, as of 10/2026)
Documentation 'Data Handling': 'No third-party sub-processors are engaged for AI Model Hub'; model developers only supply the model weights and do not process personal data on the customer's behalf
Provider’s list
Ollama
Free
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented
Pro
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented
Max
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented
Team (Early Access)
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented
Enterprise
NVIDIA Cloud Providers (NCPs) for hosting open models, without names or locations (pricing page)
Stripe (payment processing, according to the Terms and Privacy Policy)
Cloud infrastructure and model inference providers according to the Privacy Policy, without names
Sub-processor list: not publicly documented

Retention

differs by tool: IONOS AI Model HubDocumentation 'Data Handling': the…only plan; OllamaPrivacy Policyall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)Documentation 'Data Handling': the content of prompts and outputs is not logged and not stored persistently, only held in host and GPU memory during processing; the API is stateless. Metadata per call (timestamp, model, token count) is recorded for billing and operation; operational logs are retained 'in accordance with IONOS CLOUD's internal retention policy' (period not quantified). Data protection package section 5.7: after processing ends, deletion or return at the customer's choice.
Ollama
FreePrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.
ProPrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.
MaxPrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.
Team (Early Access)Privacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.
EnterprisePrivacy Policy (March 2026): for cloud models, prompts and responses are processed only transiently and not stored beyond the time required to fulfil the request; pricing page: prompt and response data is never logged. Account information until deletion on request, billing records according to tax and accounting obligations, support communications until resolved, metadata as needed for operations, then deleted or de-identified. Specific retention periods not publicly documented.

Price

differs by tool: IONOS AI Model Hubper tokenonly plan; Ollama
0 USD… (further prices in the details)Free
20 USD/mo… (further prices in the details)Pro
100 USD/moMax
from 500 USD/moTeam (Early Access)
customEnterprise; IONOS AI Model Hub
AI Model Hub (usage-based)
Usage-based, net prices excl. VAT, offer for businesses only (price list IONOS Cloud GmbH, as of 2026-10-18), per 1 million input/output tokens e.g.: Qwen3.5-9B EUR 0.10/0.15; Llama 3.1 8B Instruct EUR 0.15/0.15; Mistral Nemo Instruct EUR 0.15/0.15; Mistral Small 24B Instruct EUR 0.10/0.30; gpt-oss-120b EUR 0.15/0.65; Llama 3.3 70B Instruct EUR 0.65/0.65; Qwen3.8-27 EUR 0.40/2.70; Qwen3.5-397B-A17B EUR 0.60/3.60; Qwen3-Coder-Next EUR 0.15/0.80; LightOnOCR 2 EUR 0.15/0.30. Images: FLUX.1 [schnell] EUR 0.0288 per image; FLUX.2 [klein] 4B EUR 0.013 per first megapixel. Embeddings from EUR 0.01 per 1 million tokens.
Ollama
Free
0 USD (pricing page). Additional usage credits at token prices per million tokens, e.g. gpt-oss:120b 0.15 USD input and 0.60 USD output; off-peak prices outside 12:00–18:00 UTC on weekdays and all day at weekends. Purchased credits expire after one year; the customer is responsible for taxes (Terms).
Pro
20 USD per month or 200 USD per year (16.67 USD per month billed annually) (pricing page); one account per person. The customer is responsible for taxes (Terms).
Max
100 USD per month (pricing page); one account per person. The customer is responsible for taxes (Terms).
Team (Early Access)
from 500 USD per month (pricing page); usage beyond the included credits is billed from a shared team balance as you go. According to the FAQ, Ollama creates a separate team account on joining. The customer is responsible for taxes (Terms).
Enterprise
Custom (pricing page: "Custom", volume usage pricing; contact via sales)
: IONOS AI Model Hubyesonly plan; Ollamayesall 5 plans; IONOS AI Model Hub
AI Model Hub (usage-based)
yes (inference API for language, code, OCR, image, embedding and reranking models from third-party developers, operated on IONOS infrastructure; OpenAI-compatible API, tool calling, RAG)
Ollama
Free
yes (cloud models with starter usage credits and access to starter models; all cloud models after buying usage credits; local models unlimited; 1 concurrent request)
Pro
yes (Free plus 60 USD of usage credits per month, larger pro models, several models concurrently, 3 concurrent requests; Fast mode "coming soon" according to the pricing page)
Max
yes (Pro plus 300 USD of usage credits per month, early access to the newest models, 10 concurrent requests)
Team (Early Access)
yes (Pro plus unlimited users, 1,000 USD of usage credits per month shared across the team, centralised billing and administration, priority support, 10 concurrent requests; shared projects "coming soon" according to the pricing page)
Enterprise
yes (Team plus model access controls, cost budgets for users and API keys, private Slack channel with dedicated support, custom security questionnaires)

Evidence

: IONOS AI Model Hub7 sources; Ollama4 sources; IONOS AI Model Hub
AI Model HubProduct page, retrieved 18 Oct 2026
Price list IONOS CLOUD GmbH (DE), section AI Model HubPricing page, retrieved 18 Oct 2026
AI Model Hub – Data HandlingDocumentation, retrieved 18 Oct 2026
AI Model Hub – European Union AI ActDocumentation, retrieved 18 Oct 2026
Terms of use for AI servicesTerms and conditions, retrieved 18 Oct 2026
Data protection package (data processing), version 7.4, as of 06/2023DPA, retrieved 18 Oct 2026
IONOS Cloud sub-contractor list, as of 10/2026Sub-processor list, retrieved 18 Oct 2026
All details on the tool page
Ollama
PricingPricing page, retrieved 18 Oct 2026
Terms of Service (Last updated: May 2026)Terms and conditions, retrieved 18 Oct 2026
Privacy Policy (Last updated: March 2026)Privacy notice, retrieved 18 Oct 2026
Docs: CloudDocumentation, retrieved 18 Oct 2026
All details on the tool page

not documentedmeans: not publicly documented, not verifiable in any provider source. It does not mean “no”.

No guarantee of accuracy, not legal advice. Each entry is valid as of the date shown. Open points are listed on the detail pages.

More comparisons

IONOS AI Model Hub compared with

Ollama compared with

Individual pages: IONOS AI Model Hub, Ollama. All tools in the category: Software development.