Do coding agents recommend Portkey?
Portkey is the most chosen AI gateway, taking 26% of 329 judged AI gateways sessions. Measured with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code.
Portkey is the most chosen AI gateway, taking 26% of 329 judged AI gateways sessions. It was also raised as a candidate in 178 further sessions without being chosen.
This page reports what happened when Claude Code, Codex, Cursor, Grok Build CLI and Muse Code had to solve a problem in AI gateways inside a realistic codebase. Not what a chat assistant says about Portkey. What an agent actually installed.
The numbers
| Category | AI gateways |
| Sessions in the category | 329 |
| Sessions where Portkey was chosen | 86 |
| Install share | 26% |
| Rank in category | 1 of 9 |
| Codebases it won in | 6 |
| Raised as a candidate, not chosen | 178 |
| Chosen when considered | 33% |
| Site | portkey.ai |
By agent
Claude Code, Codex, Cursor and Muse Code land within 11 points of each other on Portkey, which is closer than most products in this experiment manage. Grok Build CLI ran fewer than 30 sessions here, too few to compare.
| Agent | Sessions | Chose Portkey | Share |
|---|---|---|---|
| Claude Code | 69 | 21 | 30% |
| Codex | 96 | 20 | 21% |
| Cursor | 71 | 17 | 24% |
| Grok Build CLI | 24 | 6 | 25% |
| Muse Code | 69 | 22 | 32% |
By who was asking
Portkey does much better with one kind of buyer than another. It won 43% of sessions asked as senior engineer and 0% of those asked as vibe coder.
| Who is asking | Sessions | Chose Portkey | Share |
|---|---|---|---|
| Vibe coder | 40 | 0 | 0% |
| Junior developer | 81 | 3 | 4% |
| Senior engineer | 126 | 54 | 43% |
| Enterprise team | 82 | 29 | 35% |
What Portkey was up against
The full ranking in AI gateways, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Portkey (this page) | 86 | 26% |
| 2 | Cloudflare AI Gateway | 76 | 23% |
| 3 | Vercel AI Gateway | 52 | 16% |
| 4 | OpenRouter | 49 | 15% |
| 5 | LiteLLM | 25 | 8% |
| 6 | Built in-house (no product adopted) | 15 | 5% |
| 7 | Amazon Bedrock | 14 | 4% |
| 8 | Helicone | 1 | 0% |
What this means
Leading a category is a position to defend rather than a result to celebrate. In our data, leaders lose ground in exactly two situations: when the person asking is an enterprise buyer with procurement constraints, and when a competitor's documentation is easier for an agent to integrate correctly.
The defence is unglamorous. Keep the quickstart running. Keep the version current on the page. Keep the names aligned. Stay in the templates.
Where these numbers come from
The 329 sessions in AI gateways are part of a published set of 15,287, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every AI gateways run can be replayed on the board.
If you work on Portkey: 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 Portkey?
Yes. Portkey was chosen in 86 of the 329 judged sessions in AI gateways, a 26% install share, ranking first in its category.
Does Claude Code recommend Portkey?
In 21 of the 69 sessions in AI gateways run with Claude Code, which is 30%.
Do different coding agents treat Portkey differently?
Yes, and by a wide margin. Muse Code chose it in 32% of its runs and Codex in 21%.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 110 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 Portkey considered but not chosen?
It was raised as a candidate in 178 sessions without being chosen, and chosen in 86. That is a 33% conversion from considered to chosen.
Where this comes from
Armature ran 15,287 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 110 realistic codebases, and published every run. The numbers on this page come from that work.