PostgreSQL Dominates 2025: 55% Adoption Crushes MySQL as All Databases Migrate

PostgreSQL achieved 55.6% developer adoption in 2025 according to Stack Overflow’s 2025 Developer Survey (49,000+ respondents), jumping 7 percentage points from 48.7% in 2024—the largest single-year expansion in PostgreSQL history. This unprecedented surge made PostgreSQL the #1 most popular, most loved (65.5% admiration), and most wanted database for three consecutive years, while MySQL fell from 2nd place in 2023 to 4th place in 2025, overtaken by Redis. The data reveals a fundamental ecosystem shift: “all databases are migrating to PostgreSQL” as organizations seek cost savings from Oracle/MySQL migrations, AI capabilities through pgvector for vector search, and modern features like JSONB for flexible data.

The 7-Point Jump: Largest PostgreSQL Expansion in History

PostgreSQL’s 7-percentage-point increase in one year (48.7% to 55.6%) represents the largest expansion in database history, with professional developer adoption reaching 58.2% and creating an 18.6-point gap over MySQL’s 39.6%. This isn’t gradual growth—it’s ecosystem consolidation. PostgreSQL achieved triple-crown status: most popular, most loved (65.5% admiration), and most wanted (46.5% desire) for three consecutive years across Stack Overflow’s annual surveys.

The competitive landscape tells the migration story clearly. While PostgreSQL surged +7 percentage points, competitors stagnated or declined: MongoDB (-0.7%), Oracle (+0.1%), and MySQL (+0.2%) showed minimal growth. MySQL’s decline accelerated from 2nd place in 2023 to 3rd in 2024 to 4th in 2025, overtaken by Redis.

JetBrains’ State of Developer Ecosystem 2025 (24,534 developers across 194 countries) confirms the trend: PostgreSQL dominates professional developer preferences across languages and regions, with AI-using professionals adopting PostgreSQL at even higher rates (59.5%). This data answers “which database should I choose for new projects?” definitively—PostgreSQL is now the default choice unless specific MySQL features are required.

AI Use Cases Drive Adoption: pgvector Powers Vector Search

PostgreSQL’s pgvector extension enables in-database vector search for AI applications—embeddings, RAG systems, semantic search—eliminating the need for separate vector databases. Version 0.8.0 achieved 9x faster query processing with binary quantization reducing memory footprint by 32x while maintaining 95% accuracy, making PostgreSQL production-ready for LLM-powered applications. Organizations using pgvector report 40-60% lower total cost of ownership compared to dedicated vector databases for datasets under 500 million vectors.

30% of Supabase’s new signups are AI builders leveraging pgvector for production workloads: chatbots, copilots, recommendation systems, fraud detection, and image search. The technical advantages are concrete—HNSW and IVFFlat indexes enable approximate nearest neighbor search at scale, while PostgreSQL’s ACID guarantees ensure data consistency that dedicated vector databases often lack.

-- Create table with vector embeddings column
CREATE TABLE documents (
  id SERIAL PRIMARY KEY,
  content TEXT,
  embedding vector(1536)  -- OpenAI text-embedding-3 dimensions
);

-- Vector similarity search using cosine distance
SELECT id, content
FROM documents
ORDER BY embedding <-> query_embedding
LIMIT 10;

-- Create HNSW index for fast approximate nearest neighbor search
CREATE INDEX ON documents USING hnsw (embedding vector_cosine_ops);

Developers building LLM applications no longer need Pinecone, Weaviate, or Milvus for vector storage—PostgreSQL handles vectors + structured data + ACID transactions in one system. This architectural simplification (one database instead of two) drives both cost savings and engineering velocity improvements, explaining why AI is a primary driver of the “all databases migrating to PostgreSQL” trend.

Enterprise Migrations: 30-40% TCO Savings from Oracle Exodus

Enterprises are actively migrating workloads from Oracle to PostgreSQL to achieve 30-40% total cost of ownership savings, avoid vendor lock-in, and gain open-source flexibility. Organizations can complete Oracle migrations in 20 days using Ora2Pg for schema conversion and SharePlex for near-zero downtime replication, making the switch from expensive Oracle licensing to open-source PostgreSQL economically compelling.

The migration economics are straightforward: Oracle’s proprietary licensing fees disappear, cloud infrastructure costs drop 30-40%, and open-source flexibility enables rapid innovation without vendor approval. Ora2Pg (free, open-source) converts Oracle schemas and data to PostgreSQL, handling the technical heavy lifting of transforming PL/SQL to PL/pgSQL and mapping proprietary Oracle data types.

MySQL to PostgreSQL migrations follow similar patterns, driven by PostgreSQL’s advanced features—JSONB for modern app architectures, parallel execution for performance, pgvector for AI workloads—that MySQL lacks or implements less robustly. One developer summarized the shift: “PostgreSQL is the more robust and versatile choice for applications with complex transactional logic, mixed transactional and analytical requirements, and the need for predictable performance under high contention.”

-- JSONB enables hybrid data model: structured + flexible
CREATE TABLE users (
  id SERIAL PRIMARY KEY,
  email TEXT,
  preferences JSONB
);

-- Create GIN index for efficient JSONB queries
CREATE INDEX ON users USING GIN (preferences);

-- Query JSONB using containment operator
SELECT * FROM users
WHERE preferences @> '{"theme": "dark", "notifications": true}';

Performance Advantages: 13x Faster Under Concurrent Load

PostgreSQL delivers 1.6x faster performance than MySQL across most operations according to academic performance benchmarks, with dramatic advantages for complex queries (0.6-0.8ms vs 9-12ms for 1 million records) and concurrent workloads. Under simultaneous operations, PostgreSQL maintains stable performance (0.7-0.9ms for select queries during inserts), while MySQL’s performance degrades significantly (7-13ms)—a 13x performance advantage that matters for high-traffic applications.

Resource efficiency compounds the advantage: PostgreSQL uses fewer system resources at equivalent loads, delivering both better performance and lower infrastructure costs. MySQL retains advantages for simple read-heavy workloads and extreme write-intensive edge cases, but these scenarios represent shrinking use cases as applications grow more complex.

The performance data explains why enterprises migrate from MySQL to PostgreSQL as applications scale. When developers ask “PostgreSQL or MySQL for my project?”, the benchmarks provide clear guidance: PostgreSQL for complex queries, concurrent workloads, and applications requiring predictable performance under load; MySQL only when extreme write intensity or legacy compatibility demands it.

Ecosystem Growth: Supabase $5B Valuation Signals Platform Strength

PostgreSQL’s ecosystem strength validates its dominance. Supabase achieved a $5 billion valuation in October 2025 (up from $2 billion in April 2025) with 4 million developers (up from 1 million in 2024), creating 2,500 new databases daily. This 300% developer growth in one year demonstrates PostgreSQL’s evolution from database to platform—mirroring how Linux fragmented into distributions (Ubuntu, Red Hat, SUSE) while becoming the default server OS.

Complementary tools reinforce the ecosystem: DuckDB (146% annual growth) provides analytics capabilities integrated with PostgreSQL, while Neon delivers serverless PostgreSQL with database branching for modern development workflows. The emergence of PostgreSQL distributions—Supabase, Pigsty, EDB—suggests the “Linux of databases” comparison isn’t hyperbole but predictive pattern recognition.

Strong ecosystem growth signals long-term viability for developers choosing databases. PostgreSQL’s surrounding infrastructure (managed hosting, analytics tools, AI extensions) is maturing rapidly, validated by venture capital ($5B Supabase valuation) and developer adoption (4M Supabase users). As one analysis noted: “PostgreSQL is becoming the default choice for databases across organizations,” with ecosystem concentration driving network effects that make alternatives less attractive over time.

Key Takeaways

• PostgreSQL hit 55.6% adoption in 2025—a 7-percentage-point jump representing the largest expansion in database history. Among professional developers, adoption reached 58.2%, creating an 18.6-point gap over MySQL’s 39.6%. MySQL declined from 2nd place (2023) to 4th place (2025), overtaken by Redis.
• AI use cases drive migration: pgvector enables in-database vector search with 9x performance improvements (v0.8.0) and 40-60% lower TCO vs dedicated vector databases. 30% of Supabase’s new signups are AI builders using pgvector for production workloads.
• Enterprise migrations from Oracle deliver 30-40% TCO savings using Ora2Pg (20-day timelines, near-zero downtime). Economic drivers—avoiding licensing fees, escaping vendor lock-in—accelerate the “all databases migrating to PostgreSQL” trend.
• Performance advantages are measurable: PostgreSQL is 13x faster for concurrent operations (0.7-0.9ms vs MySQL’s 7-13ms degradation) and 1.6x faster overall. Predictable performance under load makes PostgreSQL the robust choice for complex applications.
• Ecosystem strength validates dominance: Supabase’s $5B valuation, 4M developers, and 300% growth demonstrate PostgreSQL evolving into a platform. For developers building new projects, PostgreSQL is the default choice unless specific MySQL features are required—and PostgreSQL skills are increasingly valuable as 55%+ of the market adopts it.