MLS-C01 Is Retired: What Replaced AWS’s ML Specialty (and What to Take Now)
Short answer: the AWS Certified Machine Learning – Specialty (MLS-C01) retired on March 31, 2026. You can’t book it anymore. If you were planning to take it, or you’re finding old study materials and wondering what’s current, the exam most people should take instead is the AWS Certified Machine Learning Engineer – Associate. In English that now means the MLA-C02 beta (booked as ME1-C02: $75, 85 questions, 170 minutes), because MLA-C01’s last English exam day was September 28, 2026. If you already hold the Specialty, relax: your certification stays valid for its full 3 years from when you earned it.
What happened to MLS-C01
The ML Specialty had a long run as AWS’s flagship machine-learning credential, but AWS restructured its AI/ML certification line around the way the field actually works now. In its place sits a family of credentials at different levels: the AI Practitioner (AIF-C01) at the foundational level, the ML Engineer Associate (MLA-C01, now being replaced by MLA-C02) as the hands-on practitioner credential, and professional-level generative AI certification above that. The last day to sit MLS-C01 was March 31, 2026; certifications already earned remain active for their normal 3-year term.
MLS-C01 vs MLA-C01 at a glance
| MLS-C01 (retired) | MLA-C01 (English ended Sept 28, 2026) | |
|---|---|---|
| Status | Last exam day March 31, 2026 | English ended September 28, 2026 (English test-takers now take the MLA-C02 beta, booked as ME1-C02: $75, 85 questions, 170 min); Japanese, Korean and Simplified Chinese until MLA-C02 is generally available |
| Level | Specialty | Associate |
| Price | $300 | $150 |
| Format | 65 questions | 65 questions (50 scored), 130 min |
| Passing score | 750 / 1000 | 720 / 1000 |
| Emphasis | ML theory + modeling depth | Production ML engineering on SageMaker |
| Validity | 3 years (existing holders unaffected) | 3 years |
The real difference: theory vs production
The Specialty leaned academic — algorithm selection, statistics, modeling nuance. The ML Engineer Associate tests what ML jobs actually involve day to day: data preparation (ingestion, Glue, Data Wrangler, Feature Store — 28% of the exam), model development (training jobs, tuning, Clarify, metrics — 26%), deployment and orchestration (endpoints, pipelines, CI/CD — 22%), and monitoring and security (drift, Model Monitor, IAM — 24%). Less “which algorithm minimizes variance,” more “ship it, monitor it, keep it secure.” For most working ML practitioners, that’s a better match — and at $150 versus the Specialty’s $300, it’s half the price. Retakes, the passing score and the MLA-C02 beta price are in our AWS ML Engineer exam cost guide.
What to do, by situation
- You were studying for MLS-C01: your ML fundamentals transfer, but re-aim at the ML Engineer Associate blueprint, which weights data preparation, deployment and operations far more heavily than modeling theory. In English that means MLA-C02: start with the free MLA-C02 study guide (or the MLA-C01 study guide if you test in Japanese, Korean or Simplified Chinese), then work through the 6-week AWS ML Engineer study plan.
- You hold the ML Specialty: it’s valid until its printed expiration, but it can’t be renewed: AWS allowed recertifying by retaking MLS-C01 only until March 31, 2026. When it expires, the ML Engineer Associate (MLA-C02 in English) is the natural next credential for engineers.
- You’re earlier in the journey: if a 130-minute associate exam sounds like a lot, the AWS AI Practitioner (AIF-C01) is the gentler on-ramp ($100, no coding) and stacks naturally toward the ML Engineer Associate later.
- Buying study materials: anything targeting MLS-C01 is now studying for an exam that no longer exists. Check the exam code before you spend. MLA-C01 has already closed in English too, so read what changes in MLA-C02 before you buy anything for it.
Prep for MLA-C02 (or MLA-C01 outside English)
Free, original practice questions with plain-English explanations, no sign-up needed. MLA-C01’s last English exam day was September 28, 2026; English test-takers now sit the MLA-C02 beta.
Keep reading: Is the AWS AI Practitioner worth it? · AI-900 vs AI-901: Microsoft’s version of this same transition · Best AI certifications in 2026
HOW TO // AI is not affiliated with or endorsed by Amazon Web Services. AWS Certified Machine Learning – Specialty, MLS-C01, AWS Certified Machine Learning Engineer – Associate, and MLA-C01 are certifications of Amazon.com, Inc. or its affiliates; we reference them descriptively. Retirement details per AWS’s published certification updates; verify current policy with AWS.
Get AI Cert Watch: an email when AI certification exams change
We check the official pages of 80+ AI exams every day. When a price, version, format or retirement date changes, you’ll get a short email. At most one a week, plus the HOW TO // AI newsletter on Thursdays.
Unsubscribe anytime. Privacy policy.
Related guides
- AWS ML Engineer Associate (MLA-C01) Study Guide & Cheat Sheet
- Free AWS Machine Learning Engineer – Associate Practice Questions
- MLA-C02 Exam Guide (ME1-C02): What Changed From MLA-C01
- Is the AWS Machine Learning Engineer Associate Worth It in 2026?
- How to Pass the AWS ML Engineer Associate (MLA-C01/C02): A 6-Week Plan
Frequently asked questions
Is the AWS Machine Learning Specialty (MLS-C01) retired?
Yes. The last day to take MLS-C01 was March 31, 2026. Certifications earned before retirement remain active for their full 3-year term.
What replaced the MLS-C01 exam?
AWS restructured its AI/ML line into a family: the AI Practitioner (AIF-C01) at the foundational level, the Machine Learning Engineer - Associate as the hands-on practitioner credential, and the Generative AI Developer - Professional (AIP-C01) above that. For most former MLS-C01 candidates the ML Engineer Associate is the exam to take; in English that is now the MLA-C02 beta, because MLA-C01 ended in English on September 28, 2026.
Is the ML Engineer Associate easier than MLS-C01 was?
It is more accessible: associate level rather than specialty, $150 for MLA-C01 or $75 for the English MLA-C02 beta rather than $300, and focused on production ML engineering rather than modeling theory. It is still a serious exam: MLA-C01 has 65 questions in 130 minutes and the MLA-C02 beta 85 questions in 170 minutes, both with 720 out of 1,000 to pass.
Is my existing MLS-C01 certification still valid?
Yes. It stays valid until the expiration printed on it (3 years from when you earned it), but it can no longer be renewed, because AWS allowed recertifying by retake only until March 31, 2026. When it expires, the ML Engineer Associate is the natural next credential for ML engineers.
