Free to start

Privatemode

Try Privatemode on us. Starter credit covers your first requests, then one key reaches all 7 Privatemode models, with the privacy posture shown on every route.

Privatemodeprivatemode

Trusted Confidential ZDR

All providers

These privacy labels describe Privatemode, the upstream model provider. Confidential requires all three: verified provider compute, end-to-end encryption and explicit zero data retention. Account and billing metadata are outside the content-retention claim.

Trusted adds TrustedRouter's assessment of the operator's physical-security posture and reliability history. Tinfoil and Privatemode are the only operators in this tier.

Provider	Privatemode
API operator country	DE Legal home of the API operator, not an inference-location guarantee.
Headquarters	Not verified
Routing status	Active
Provider website	https://www.privatemode.ai/
Models	7 public models
Credits routes	7
Zero data retention	yes All routes
Verified confidential inference	Verified
Policy note	Requests and responses are encrypted between TrustedRouter's enclave and release-pinned Privatemode workloads after attestation verification. Verification failures reject the request; there is no plaintext fallback. Privatemode explicitly states that prompts and responses are not stored after a request completes and are not used for training. Transient inference state may remain in a salt-isolated, in-memory prompt cache inside confidential workers until eviction. Operational metadata is retained for up to 90 days; per-key token usage is retained permanently for billing. Cache-routing metadata is visible to the provider edge. ZDR describes prompt/output retention, not metadata. Policy source

Inference locations

Provider API declarations are also available on each model's endpoint API. Direct chat and Responses replies separate advertised availability from serving location in usage.inference_location; the actual serving region remains unknown when not reported.

usage.provider_usage.gateway_region identifies the TrustedRouter gateway, not the provider's inference location. A value such as europe-west4 does not mean the model ran in Europe. usage.inference_location.serving_region remains null when the provider has not reported it. Older gateways may return the gateway-only field as provider_usage.region.

Where the upstream GPUs execute the model, not the provider's headquarters, API ingress, or data storage. ZDR and confidential inference do not by themselves restrict geography.

Declared countries or regions: European Union (countries unspecified)
Privatemode declares EU hosting for its inference service. This does not establish an individual worker's country.
Can locations change?: Changes within the EU and exact model placement are not specified by the reviewed declaration.
Provider region pinning: No per-request country-selection contract verified.
Pinning through TrustedRouter: Not supported by the current TrustedRouter integration.
Infrastructure declaration: EU-hosted inference, provider-declared; country and individual worker location unspecified.
Evidence: Public provider statement; encryption attestation is separate from location evidence. Reviewed 2026-09-27.

Privatemode EU hosting

Provider selection alone is not a region pin. TrustedRouter's US provider filter selects company jurisdiction, not GPU location. A regional gateway hostname does not pin a downstream GPU. For a binding residency requirement, confirm the model, region, failover scope and enforced controls before sending sensitive data.

Measured performance

60 samples

Continuously sampled across Privatemode's routed models: p50 TTFT, effective throughput, and success rate. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime. No prompt or output content stored.

p50 TTFT	1323 ms
Effective throughput	42 tok/s n=2
Uptime	100.00%

Model	p50 TTFT	Effective throughput	Uptime	Config excluded	Availability samples
z-ai/glm-5.3-flash	1323 ms	—	100.00%	—	39
openai/gpt-oss-120b	958 ms	—	100.00%	—	7
deepseek/deepseek-ocr-2	1102 ms	—	100.00%	—	9
z-ai/glm-5.3	1237 ms	42 tok/s n=2	100.00%	1 `router_error`	5

Privatemode performance history · Full provider & model leaderboard.

Provider models

Models served by Privatemode.

Each row links to pricing, provider, benchmark, and API pages for the model.

Model	AI IQ	Context	Input	Cached input	Output
`deepseek/deepseek-ocr-2` DeepSeek OCR 2	—	8,192	$0.949097/1M	$0.098608/1M	$1.565395/1M
`mistralai/voxtral-mini-3b-2507` Voxtral Mini 3B	—	7 MiB, 10 minutes	$0.004932/audio minute	Not applicable	Included
`openai/gpt-oss-120b` OpenAI: gpt-oss-120b	IQ 98#136	131,072	$0.246519/1M	$0.049305/1M	$0.801186/1M
`openai/whisper-large-v3` Whisper Large V3	—	7 MiB, 10 minutes	$0.017257/audio minute	Not applicable	Included
`qwen/qwen3-embedding-4b` Qwen3-Embedding 4B	—	32,768	$0.160238/1M	Not published	selected route
`z-ai/glm-5.3` Z.ai: GLM 5.3	IQ 124#30	1,048,576	$1.910521/1M	$0.18489/1M	$7.075087/1M
`z-ai/glm-5.3-flash` Z.ai: GLM 5.3 Flash	IQ 119#50	1,048,576	$0.246519/1M	$0.06163/1M	$0.801186/1M

Questions

Does Privatemode have zero data retention?

TrustedRouter records Privatemode as supporting provider-level zero data retention based on the policy source linked on this page. This is a provider policy claim, separate from TrustedRouter's content-stateless real-time gateway and from end-to-end confidential compute.

Is Privatemode end-to-end encrypted?

TrustedRouter records Privatemode as supporting provider-side confidential compute and end-to-end encrypted inference. The route-specific model page shows whether that protection applies to a particular endpoint.

Which Privatemode models are available through TrustedRouter?

This page currently lists 7 public Privatemode models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.