Do coding agents recommend OpenAI Vector Stores?
OpenAI Vector Stores was chosen in 15% of 158 judged vector search sessions, ranking second. Measured with Claude Code, Codex and Cursor.
OpenAI Vector Stores was chosen in 15% of 158 judged vector search sessions, ranking second. It was also raised as a candidate in 1 further sessions without being chosen.
This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in vector search inside a realistic codebase. Not what a chat assistant says about OpenAI Vector Stores. What an agent actually installed.
One thing to read first: every one of those wins came from a single codebase. That is a result about one repository rather than about vector search in general, and the splits below cannot separate the two. Treat them as a description of that repository.
The numbers
| Category | Vector search |
| Sessions in the category | 158 |
| Sessions where OpenAI Vector Stores was chosen | 24 |
| Install share | 15% |
| Rank in category | 2 of 15 |
| Codebases it won in | 1 |
| Raised as a candidate, not chosen | 1 |
| Chosen when considered | 96% |
| Site | openai.com |
By agent
The three agents land within 8 points of each other on OpenAI Vector Stores, which is closer than most products in this experiment manage.
| Agent | Sessions | Chose OpenAI Vector Stores | Share |
|---|---|---|---|
| Claude Code | 58 | 6 | 10% |
| Codex | 66 | 12 | 18% |
| Cursor | 34 | 6 | 18% |
By who was asking
OpenAI Vector Stores does much better with one kind of buyer than another. It won 40% of sessions asked as junior developer and 0% of those asked as enterprise team.
| Who is asking | Sessions | Chose OpenAI Vector Stores | Share |
|---|---|---|---|
| Vibe coder | 45 | 0 | 0% |
| Junior developer | 60 | 24 | 40% |
| Senior engineer | 26 | 0 | 0% |
| Enterprise team | 27 | 0 | 0% |
What OpenAI Vector Stores was up against
The full ranking in vector search, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Neon | 54 | 34% |
| 2 | OpenAI Vector Stores (this page) | 24 | 15% |
| 3 | Qdrant | 23 | 15% |
| 4 | Typesense | 20 | 13% |
| 5 | Pinecone | 19 | 12% |
| 6 | Meilisearch | 7 | 4% |
| 7 | Elasticsearch | 2 | 1% |
| 8 | turbopuffer | 2 | 1% |
What this means
A solid second or third position in a category means the agent is genuinely choosing rather than reaching automatically. That is a winnable position, because the inputs it uses can be changed.
The fastest gains are usually in the sessions that were nearly won: read them, find the step where the agent moved on, and fix that step.
Where these numbers come from
The 158 sessions in vector search are part of a published set of 5,915, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every vector search run can be replayed on the board.
If you work on OpenAI Vector Stores: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
Do coding agents recommend OpenAI Vector Stores?
Yes. OpenAI Vector Stores was chosen in 24 of the 158 judged sessions in vector search, a 15% install share, ranking second in its category.
Does Claude Code recommend OpenAI Vector Stores?
In 6 of the 58 sessions in vector search run with Claude Code, which is 10%.
Do different coding agents treat OpenAI Vector Stores differently?
Not much. The three agents chose it at similar rates, between 10% and 18% of their runs.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 56 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.
How often is OpenAI Vector Stores considered but not chosen?
It was raised as a candidate in 1 sessions without being chosen, and chosen in 24. That is a 96% conversion from considered to chosen.
Where this comes from
Armature ran 5,915 judged sessions with Claude Code, Codex and Cursor inside 56 realistic codebases, and published every run. The numbers on this page come from that work.