micro1 vs Mercor
micro1 vs Mercor for AI trainers: listed pay, interview length, payment schedule, work stability and who should apply to which.
How micro1 vs Mercor compares
| Attribute | MiMicro1 | MeMercor |
|---|---|---|
| Pay and payment | ||
| Median advertised rate | $60/hr across 401 listings with a usable rate. The middle half of listings run $40 to $90/hr. | Better here$77.50/hr across 1,244 listings. The middle half run $50 to $100/hr. (Census of job listings, 8 Sept 2026.) |
| Rates for senior experts | Most roles pay $40 to $90/hr. A tail reaches about $150/hr for senior engineering and specialist work. One guide reports up to $500/hr for rare medical and finance specialisms. | Better hereCommonly cited at $80 to $150/hr, with $200+/hr at the top for senior domain experts. One guide estimates Mercor pays top experts 30 to 50% more for equivalent work. |
| Payment schedule | Sources disagree. An October 2026 guide says twice a month. A May 2026 guide says weekly. Check what your contract says. | Better hereWeekly. The pay week runs Saturday to Friday and money arrives around the following Wednesday. One review says the first Stripe payout is held for 7 days. |
| Payment provider | Deel, according to one guide. Another says Wise or Stripe. Identity verification and a background check come before your first payment. | Stripe, or Wise where Stripe is not supported. |
| Fees charged to workers | None reported. | None under Mercor's published policy. |
| Getting in | ||
| Interview | Better hereAn adaptive AI interview called Zara. Reports put it at 15 to 30 minutes, with live coding for engineers. | An AI-led video interview. Accounts range from about 20 minutes to 90 minutes, and one says you can retake it. |
| How selective it is | Better hereLenient. One guide says most applicants with verifiable credentials get through, and strong writers without a specialty can place. | Strict. One guide says most applicants are rejected and generalists rarely place. Credentials like a PhD, an MD, a JD or five years at a top firm help. |
| Wait after you pass | Better hereYou apply to specific roles yourself, so the wait depends on how many open roles fit you. | You are matched to projects. Reported waits for a first engagement run from two to six weeks. |
| Work and stability | ||
| How you find projects | You browse open roles and apply. | Matching is algorithmic and, according to one guide, you generally cannot pick engagements. |
| Typical work | Code evaluation in Python, TypeScript, Go and Rust, RLHF preference labeling, analyst-style review, agent training and robotics data programs. | Expert evaluation and red-teaming for frontier labs and enterprises: medical reasoning, legal analysis, graduate-level math and senior code review. |
| Work stability | Better hereDescribed as more continuous. Some contractors stay with one client for months. | Often uneven. Engagements of two to six weeks followed by gaps are common, per one guide. |
| Time tracking | Depends on the project: hourly, per task or flat fee. | Desktop app with periodic screenshots on hourly work. One review says no manual time entry and a weekly billable-hour cap. |
| Working for both | No exclusivity clause reported. | No exclusivity clause reported. |
| Trust and risk | ||
| Public reviews | Better hereTrustpilot about 4.7/5 in August 2026, up from about 2.2 in October 2025. Glassdoor 4.5/5. | Trustpilot 4.1/5 from about 226 reviews in August 2026. |
| Security incident | Better hereNone reported in the sources we checked. | Disclosed a data breach on 31 March 2026, traced to a compromised LiteLLM package. Reports say up to 4 TB of internal data was exposed, including contractor personal information. Several contractors have sued. Those claims are allegations. |
| Most common complaint | Being accepted and then waiting for work. | The same wait for work. Contractors also told The Verge, New York Magazine and Wired about stressful management and declining pay. |
| Company | ||
| Funding | $35M Series A at a $500M valuation (September 2025). Forbes reported a later round of more than $100M at about $4B, citing people familiar with the deal. | $350M Series C at a $10B valuation (October 2025). |
| Scale | Reported $500M gross run rate in August 2026. | Reported $2B gross annualised run rate in July 2026. About 30,000 contractors as of October 2025. |
Which one should you pick
If you hold senior credentials, such as a PhD, an MD, a JD or years at a top firm, apply to Mercor first. That is where the highest rates show up, and the listing data backs it: a $17.50 an hour gap between the two medians. If you are earlier in your career, want more say over which projects you take, or want a better shot at getting through screening, start with micro1. Do not plan around the top-end figures. They are advertised rates, and the most common complaint on both platforms is passing the screen and then waiting for work. If you are careful about where your ID and personal details sit, weigh Mercor's March 2026 breach before you upload anything. One guide reports that neither platform has an exclusivity clause. If your contracts agree, applying to both is the cheapest way to find out which one sends you work.
Pros and cons
Apply links may be referral links. They cost you nothing and help keep this site running.
Frequently asked questions
Sources
Pay figures are advertised or reported rates and vary by domain, country and project. Check the platform before you commit time to an assessment.
- aitrainer.work: pay rate benchmark of active listings (census 8 Sep 2026)
- getAIwork: Mercor vs Micro1 (updated 4 Oct 2026)
- AI Training Jobs Finder: Mercor vs micro1 (17 May 2026)
- Para Labs: Alignerr vs Mercor vs Outlier (Sep to Oct 2026)
- Wikipedia: Mercor
- TechCrunch: micro1 raises at a $500M valuation (12 Sep 2025)
- Forbes Australia: micro1 reported $4B valuation