Mercor vs Turing: Expert Ceiling vs Steady Pipeline

Mercor and Turing both place specialists into AI training work, and they optimize for different things. Mercor chases the top of the market and pays for it. Turing runs a wider, more standardized pipeline with a lower median and a stricter schedule. The choice is less about prestige than about which constraint you can live with: rate volatility or slow withdrawals.

Pay: a higher ceiling versus a tighter corridor#

Mercor advertises $50–$150/hr on its expert track, with an average above $85/hr. Generalist work sits between $16 and $100+/hr, and fixed project amounts appear alongside hourly roles. The band is wide because Mercor prices the credential and the client, not a standard track.

Turing standardizes pay. A July 2026 benchmark of 2,628 AI training listings put its median at about $27/hr, one of the narrowest corridors in the market. Expert tracks still pay well: finance experts reach $80–$150/hr and medical or science experts $70–$110/hr. The generalist median is the number most applicants actually see.

Screening and time to work#

Mercor moves faster on paper. The screen is an AI video interview followed by a trial project at a lower rate. In practice, the trial is the work sample, and project assignment still depends on client demand.

Turing asks for a 45-minute analytical challenge, then a 30-minute writing assessment for non-coders or LeetCode-style tests for developers. Onboarding takes 5–14 days. Contracts run one to three months and require a daily 2–5 hour overlap with US Pacific time, which rules out some applicants before pay even enters the conversation.

Payment speed is the real differentiator#

Mercor pays weekly. The documented complaint is about rate cuts and project transitions, not missing payments. Turing pays biweekly or monthly in USD, and the recurring 2026 complaint is withdrawal delays of weeks to months. Reviewers describe a payments-operations problem rather than a scam. That distinction matters: the money is documented, but the timing is not reliable enough to run a household on.

Who should pick which#

Pick Mercor if you hold a strong credential, want the highest expert ceiling, and can treat each project as a short engagement. Pick Turing if you want a steadier flow of contracts, can work US Pacific hours, and can carry a slow payment cycle. A finance or medical expert can do well on either; a generalist will feel the Turing median more than the Mercor ceiling.

The two also differ in eligibility. Mercor publishes no US-only requirement, though individual roles can restrict location. Turing operates in multiple markets, but the Pacific overlap requirement narrows who can take the work. Check the listing before you invest in an assessment.

Written for real AI training and domain expert candidates. No fluff, no recycled job board advice.