Zro

Private inference for coding agents

API

Developer Tools

Tech

Vercel Day

Visit WebsiteSee on Product HuntTwitter ⧉

"Zero request retention" as a verifiable claim is the hard part of private inference, since the whole value proposition depends on users trusting that it's true without being able to audit it themselves. Is there a technical mechanism that makes this verifiable, like a transparency report, third-party audit, or something architectural like processing in a TEE that provides cryptographic guarantees, or is it currently a policy commitment rather than a technical one?

@Emir Soytürk makes sense, TTFT and TPS are the two numbers that actually matter for an agent loop, not just raw throughput. curious if you've published any benchmarks against the big hosted providers or if that's still coming

Congrats on the launch! Zero retention really matters once you're touching more than one company's data, I run a few different businesses through my own tool and the idea of any of that context leaking across clients or getting used for training is a non-starter for me. Curious how you think about audit logs on your end, if a customer asks "prove nothing was retained," what do you actually show them?

👋 Hey Product Hunt!

Quick correction: our Product Hunt launch offer is no longer 100% off.

We saw abuse from zero-dollar checkout flows, so we changed the Pro launch offer to first month for $1. It is still a major launch discount, but it requires a real payment and helps us keep the service healthy for legitimate users.

The Max Plan promo code is unchanged: 50% off the first month.

Thanks for understanding. We’d rather be transparent and keep things stable for everyone.

zero retention is the right pitch for coding agents specifically, a lot of codebases can't go through a provider that might log or train on the prompts. is the latency comparable to the big hosted providers, or is there a tradeoff for the privacy guarantee

Do you also support Claude? Would be interested if you supported their models.

Private inference is one of the most underrated problems in the agent space right now. I build in a compliance-heavy vertical, so "your code never leaves your control" is a real buying trigger, not a nice-to-have.

How are you handling the latency tradeoff vs. hosted frontier models?

Streaming support for token-by-token responses would be huge for chat use cases, even if it means slightly more work on the caching layer. Right now waiting for full completions before anything renders feels dated for an inference API in 2026.

private + coding agents, finally. how much latency do you give up to keep it private?

Congrats on the launch! Keeping it OpenAI-compatible is a massive smart move. A lot of teams want to switch to open-weight models for privacy but dread rewriting their entire agent orchestration layer or dealing with completely different API schemas. How has the drop-in replacement experience been for early testers? Are there any specific edge cases with function calling or streaming where the compatibility layer behaves differently than native OpenAI?

Private inference is the box teams check last, usually right after their proprietary code turns up somewhere it should not. The part I would pressure-test is latency ten tool calls deep in a real agent loop, not on a single prompt. That is where most self-hosted setups fall over.

I am using GLM 5.2 through ZRO in my daily workflow for a while now. The speed and quality have been impressive so far.

Good luck with the launch!

the zero-retention, no-training-on-your-code angle is the whole pitch for me. most devs just assume their prompts end up as training data somewhere and live with it. if you can actually prove retention is zero, that's a real reason to switch, not just a nice-to-have

honestly the no-retention thing is a really thoughtful call, most teams talk about speed but skip the privacy piece entirely. nice to see it baked in from day one

I've been using Zro for about a month, it's great!
Very fast and reliable

Really like the commitment to zero request retention, that's a rare stance these days and shows the team actually thought through the privacy side instead of just slapping it on a landing page.

"Your code shouldn't be someone else's training data" is a strong line. Which open models are you running under the hood, and is EU hosting the default or a paid tier? Congrats on the launch

The API speed of response is really fast. Been used for a while and very satisfied by the quality of the API.

Love that there's no request retention here, that's a real differentiator for anyone dealing with sensitive data. One thing that would help me evaluate it faster though: a small public latency dashboard comparing Zro to other inference providers on common open models. Even just a simple weekly update with p50 and p95 numbers across regions would make it way easier to decide if it's worth migrating workloads over.