Do coding agents recommend Perplexity Sonar?
Perplexity Sonar was chosen in 3% of 277 judged AI search sessions, ranking ninth. Measured with Claude Code, Codex and Cursor.
Perplexity Sonar was chosen in 3% of 277 judged AI search sessions, ranking ninth. It was also raised as a candidate in 99 further sessions without being chosen.
This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in AI search inside a realistic codebase. Not what a chat assistant says about Perplexity Sonar. What an agent actually installed.
The numbers
| Category | AI search |
| Sessions in the category | 277 |
| Sessions where Perplexity Sonar was chosen | 9 |
| Install share | 3% |
| Rank in category | 9 of 27 |
| Codebases it won in | 3 |
| Raised as a candidate, not chosen | 99 |
| Chosen when considered | 8% |
| Site | perplexity.ai |
By agent
With 9 wins spread across three agents, the rates below are small numbers and a difference between them is not yet a finding. They are here because the direction is worth knowing, not because the gap is established.
| Agent | Sessions | Chose Perplexity Sonar | Share |
|---|---|---|---|
| Claude Code | 92 | 0 | 0% |
| Codex | 96 | 5 | 5% |
| Cursor | 89 | 4 | 4% |
By who was asking
Perplexity Sonar performs similarly across the four kinds of buyer, from 0% to 14%. That is unusual: the category leader changed with the persona in 20 of the 25 categories we measured.
| Who is asking | Sessions | Chose Perplexity Sonar | Share |
|---|---|---|---|
| Vibe coder | 44 | 6 | 14% |
| Junior developer | 117 | 2 | 2% |
| Senior engineer | 68 | 1 | 1% |
| Enterprise team | 48 | 0 | 0% |
What Perplexity Sonar was up against
The full ranking in AI search, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Anthropic web search | 56 | 20% |
| 2 | Exa | 38 | 14% |
| 3 | Brave Search API | 31 | 11% |
| 4 | Tavily | 21 | 8% |
| 5 | OpenAI web search | 16 | 6% |
| 6 | Firecrawl | 14 | 5% |
| 7 | Parallel | 13 | 5% |
| 8 | Built in-house (no product adopted) | 13 | 5% |
What this means
This is an integration problem, not a presence problem. Agents raised Perplexity Sonar in 99 sessions and chose it in 9, so it reaches the shortlist and then loses. Something at the last step is costing the session, and in our data that is usually a quickstart that does not run when pasted, documentation describing an interface that changed, or a package name that does not match the product name.
That is the cheaper of the two problems to have. The reason is written down in each losing transcript.
Where these numbers come from
The 277 sessions in AI search are part of a published set of 7,025, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every AI search run can be replayed on the board.
If you work on Perplexity Sonar: 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 Perplexity Sonar?
Yes. Perplexity Sonar was chosen in 9 of the 277 judged sessions in AI search, a 3% install share, ranking ninth in its category.
Does Claude Code recommend Perplexity Sonar?
In 0 of the 92 sessions in AI search run with Claude Code, which is 0%.
Do different coding agents treat Perplexity Sonar differently?
Not much. The three agents chose it at similar rates, between 0% and 5% of their runs.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 77 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 Perplexity Sonar considered but not chosen?
It was raised as a candidate in 99 sessions without being chosen, and chosen in 9. That is a 8% conversion from considered to chosen.
Where this comes from
Armature ran 7,025 judged sessions with Claude Code, Codex and Cursor inside 77 realistic codebases, and published every run. The numbers on this page come from that work.