Tinfoil
Hacker ApprovedRanked 3 of 12 LLM inference APIs by recommenders
People recommend Tinfoil for verifiably private AI inference in secure enclaves, with open-source code and attestation. Critics note it has not been independently audited.
People recommend Tinfoil for verifiably private AI inference in secure enclaves, with open-source code and attestation. Critics note it has not been independently audited.
31 people gave an opinion
- recommend it
- 21
- criticize it
- 10
- approval share
- 68%
Recommended in 10 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 21 people recommend itNeeds 10 or more, met.
- 2.1 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 10 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Verifiable privacy via secure enclaves
“Tinfoil OTOH claims hardware attestation and confidential computing, which is a pretty strong and (in theory) verifiable promise.”
Comment from 18 Aug 2026 in “GPT-5.6 Sol Pricing Cut by 50%” “What you really want (assuming you can't afford your own servers and datacentre space) is secure enclave processing”
Comment from 19 Oct 2026 in “MiniMax M3.1 Flash” “we offer providers like Tinfoil that are also end to end that we also attested using Secure Enclaves”
Comment from 31 May 2026 in “OpenRouter raises $113M Series B” - Comment from 14 Aug 2026 in “Google Is Making Private AI Practical with Homomorphic Encryption”
- Comment from 10 May 2026 in “Local AI needs to be the norm”
- Comment from 2 May 2026 in “Show HN: Filling PDF forms with AI using client-side tool calling”
- Comment from 16 Jul 2026 in “Kimi K3 is now live”
- Comment from 22 May 2026 in “DeepSeek makes the V4 Pro price discount permanent”
- Comment from 26 Feb 2026 in “How Will OpenAI Compete?”
Open source, verifiable code
“they seem to pretty open about their tech stack which is hugely differentiating compared to every other AI company”
Comment from 27 Aug 2026 in “I miss the old Claude Code” “Tinfoil are forced to publish what they run in the enclave in their GitHub account”
Comment from 19 Oct 2026 in “MiniMax M3.1 Flash” “All the code running inside the enclave is open source”
Comment from 9 Jun 2026 in “Apple reveals new AI architecture built around Google Gemini models” - Comment from 27 Jan 2026 in “Show HN: Ourguide – OS wide task guidance system that shows you where to click”
Cloud use with local-level privacy
“using services such as Tinfoil or Privatemode is an enormous step up from "trust me dude, we won't look at your data"”
Comment from 28 Jun 2026 in “Austria Lobbies EU to Host Anthropic After US Access Curbs” “lets you use the model in the cloud while getting similar privacy as running locally”
Comment from 2 May 2026 in “DeepSeek V4–almost on the frontier, a fraction of the price” “personally (ie privacy not liability) I trust it more than a contract”
Comment from 18 Aug 2026 in “GPT-5.6 Sol Pricing Cut by 50%” Easy integration and reliable use
“I loaded credits into Tinfoil the other day and have had zero issues”
Comment from 18 Aug 2026 in “GPT-5.6 Sol Pricing Cut by 50%” “fully private and attestable AI which works amazing with Kilo Code”
Comment from 26 Aug 2026 in “I miss the old Claude Code” “tinfoil is very good, easy to integrate using their cli proxy”
Comment from 2 Apr 2026 in “Anthropic: Claude Code users hitting usage limits 'way faster than expected'”
What people warn about
Each mark is one comment
Monitoring and logging concerns
“One of them (Tinfoil) just put server-side scanning inside the enclave for the consumer product.”
Comment from 19 Oct 2026 in “MiniMax M3.1 Flash” “the tinfoil endpoint I used is heading towards surveillance as well by adding abuse monitoring”
Comment from 13 Oct 2026 in “Just say no to OpenAI and Anthropic” “It exfiltrates a log of classifier activations by ID and IP to Tinfoil.”
Comment from 20 Oct 2026 in “MiniMax M3.1 Flash” High prices and cache pricing
“tinfoil doesn't seem to distinguish between cached and input tokens. I think that would make it untenably expensive”
Comment from 24 Jun 2026 in “Unsloth GLM-5.2 – How to Run Locally” “Their prices are extremely high compared to the competition”
Comment from 18 Aug 2026 in “GPT-5.6 Sol Pricing Cut by 50%” Model list inaccurate or outdated
“Your pricing faq says “all models are listed above.” They are not.”
Comment from 23 May 2026 in “DeepSeek makes the V4 Pro price discount permanent” “the models are old (Tinfoil)”
Comment from 2 Sep 2026 in “My local model setup on an M4 Pro Mac Mini”
Other LLM inference APIs people recommend
The four most recommended other LLM inference APIs, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Cerebras Inference | 109 | 51 | 68% | 13 | — | Approved |
| Groq API | 58 | 23 | 72% | 12 | — | Approved |
| TinfoilThis page | 21 | 10 | 68% | 10 | — | Approved |
| Darkbloom | 19 | 39 | 33% | 1 | — | — |
| Synthetic | 18 | 1 | 95% | 11 | — | Approved |
What people said
40 of the 48 comments that mention it, newest first. Each links to the original comment.
“It exfiltrates a log of classifier activations by ID and IP to Tinfoil.”
“Tinfoil are forced to publish what they run in the enclave in their GitHub account”
“One of them (Tinfoil) just put server-side scanning inside the enclave for the consumer product.”
“What you really want (assuming you can't afford your own servers and datacentre space) is secure enclave processing”
“the tinfoil endpoint I used is heading towards surveillance as well by adding abuse monitoring”
“probably the only non-evil use of Trusted Computing I've ever seen”
“two high quality companies:”
“I'd point to tinfoil.sh. I'm not a shill for tinfoil, I haven't even used it or looked past its homepage really”
“Tinfoil has some newer ones too like GLM 5.3 now.”
“the models are old (Tinfoil)”
“tinfoil [dot] sh looks promising”
“TEE inference that works well”
“they seem to pretty open about their tech stack which is hugely differentiating compared to every other AI company”
“fully private and attestable AI which works amazing with Kilo Code”
“Tinfoil OTOH claims hardware attestation and confidential computing, which is a pretty strong and (in theory) verifiable promise.”
“Their prices are extremely high compared to the competition”
“I loaded credits into Tinfoil the other day and have had zero issues”
“personally (ie privacy not liability) I trust it more than a contract”
“there are already providers that are roughly 2x the cost of normal providers”
“I would much prefer to use on-prem or - at most - one of the secure-enclave providers like Tinfoil”
“throw this onto the wall of fame”
“verifiably private AI that runs in enclaves, e.g. Tinfoil (US)”
“I'm pretty excited about confidential computing and startups building on it, like Tinfoil, for some use cases”
“using services such as Tinfoil or Privatemode is an enormous step up from "trust me dude, we won't look at your data"”
“tinfoil doesn't seem to distinguish between cached and input tokens. I think that would make it untenably expensive”
“Tinfoil have not been independently audited, it is somewhere on their long-term radar.”
“All the code running inside the enclave is open source”
“we offer providers like Tinfoil that are also end to end that we also attested using Secure Enclaves”
“Your pricing faq says “all models are listed above.” They are not.”
“all models are hosted in an attested secure hardware enclave”
“this looks potentially perfect for a tool I’m making that involves financial data”
“They solved all this stuff, I'm surprised more people aren't aware of it.”
“verifiably private inference with open source models running inside secure enclaves”
“we created Tinfoil because not everyone has that capability especially when it comes to larger models”
“lets you use the model in the cloud while getting similar privacy as running locally”
“verifiably private AI inference with frontier open source models”
“tinfoil is very good, easy to integrate using their cli proxy”
“E2E encrypt all data to a secure hardware enclave”
“Tinfoil is open-source and fully verifiable.”
“This is a similar architecture to what we've been developing at Tinfoil”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.