Free to start

Tinfoil

Try Tinfoil on us. Starter credit covers your first requests, then one key reaches all 11 Tinfoil models, with the privacy posture shown on every route.

Tinfoiltinfoil

Confidential ZDR

All providers

These privacy labels describe Tinfoil, 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.

Provider	Tinfoil
API operator country	US Legal home of the API operator, not an inference-location guarantee.
Headquarters	Not verified
Routing status	Active
Provider website	https://tinfoil.sh/
Models	11 public models
Credits routes	11
Zero data retention	yes All routes
Verified confidential inference	Verified
Policy note	Tracked as a confidential inference provider with attested provider compute and no prompt/output logging claims. 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: Not verified
No inference-location declaration has been verified for this provider.
Can locations change?: Not verified. Do not assume that the serving location is fixed.
Provider region pinning: Not verified.
Pinning through TrustedRouter: No verified provider-specific location pin documented here.
Infrastructure declaration: Inference location not verified; no country-specific residency commitment.
Evidence: Not yet reviewed.

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.

Provider models

Models served by Tinfoil.

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

Model	AI IQ	Context	Input	Cached input	Output
`deepseek/deepseek-v4.1-flash` DeepSeek: DeepSeek V4.1 Flash	IQ 126#22	1,048,576	$0.68575/1M	$0.13715/1M	$1.52975/1M
`deepseek/deepseek-v4.1-flash-1017` DeepSeek: DeepSeek V4.1 Flash 1017	IQ 127#18	1,048,576	$0.68575/1M	$0.13715/1M	$1.52975/1M
`google/gemma-4-31b-it` Google: Gemma 4 31B	IQ 101#116	262,144	$0.422/1M	Not published	$1.055/1M
`meta-llama/llama-3.3-70b-instruct` Meta: Llama 3.3 70B Instruct benchmarks	—	131,072	$1.84625/1M	Not published	$2.90125/1M
`mistralai/voxtral-4b-tts-2603` Mistral: Voxtral 4B TTS	—	4096 characters	$0.00211/request	Not applicable	Included
`moonshotai/kimi-k3` MoonshotAI: Kimi K3	IQ 127#21	1,048,576	$4.22/1M	$0.844/1M	$21.1/1M
`openai/gpt-oss-120b` OpenAI: gpt-oss-120b	IQ 98#132	131,072	$0.15825/1M	Not published	$0.633/1M
`openai/gpt-oss-safeguard-120b` GPT-OSS Safeguard 120B	—	131,072	$0.15825/1M	Not published	$0.633/1M
`openai/whisper-large-v3-turbo` Whisper Large V3 Turbo	—	7 MiB, 10 minutes	$0.01055/request	Not applicable	Included
`qwen/qwen3-tts` Qwen: Qwen3 TTS	—	4096 characters	$0.01055/request	Not applicable	Included
`z-ai/glm-5.3` Z.ai: GLM 5.3	IQ 124#28	1,048,576	$1.899/1M	$0.47475/1M	$6.06625/1M

Questions

Does Tinfoil have zero data retention?

TrustedRouter records Tinfoil 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 Tinfoil end-to-end encrypted?

TrustedRouter records Tinfoil 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 Tinfoil models are available through TrustedRouter?

This page currently lists 11 public Tinfoil models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.