Do coding agents recommend Arize Phoenix?
Arize Phoenix was chosen in 7% of 478 judged evals sessions, ranking third. Measured with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code.
Arize Phoenix was chosen in 7% of 478 judged evals sessions, ranking third. It was also raised as a candidate in 131 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 evals inside a realistic codebase. Not what a chat assistant says about Arize Phoenix. What an agent actually installed.
The numbers
| Category | Evals |
| Sessions in the category | 478 |
| Sessions where Arize Phoenix was chosen | 32 |
| Install share | 7% |
| Rank in category | 3 of 11 |
| Codebases it won in | 2 |
| Raised as a candidate, not chosen | 131 |
| Chosen when considered | 20% |
| Site | arize.com |
By agent
Claude Code, Codex, Cursor and Muse Code land within 11 points of each other on Arize Phoenix, 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 Arize Phoenix | Share |
|---|---|---|---|
| Claude Code | 120 | 7 | 6% |
| Codex | 144 | 16 | 11% |
| Cursor | 120 | 7 | 6% |
| Grok Build CLI | 22 | 2 | 9% |
| Muse Code | 72 | 0 | 0% |
By who was asking
Arize Phoenix performs similarly across the four kinds of buyer, from 0% to 14%. That is unusual: the category leader changed with the persona in 24 of the 31 categories we measured.
| Who is asking | Sessions | Chose Arize Phoenix | Share |
|---|---|---|---|
| Junior developer | 241 | 0 | 0% |
| Senior engineer | 237 | 32 | 14% |
What Arize Phoenix was up against
The full ranking in evals, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Built in-house (no product adopted) | 161 | 34% |
| 2 | Langfuse | 148 | 31% |
| 3 | Promptfoo | 70 | 15% |
| 4 | Arize Phoenix (this page) | 32 | 7% |
| 5 | Braintrust | 23 | 5% |
| 6 | Inspect AI | 16 | 3% |
| 7 | Helicone | 12 | 3% |
| 8 | LangSmith | 9 | 2% |
What this means
This is an integration problem, not a presence problem. Agents raised Arize Phoenix in 131 sessions and chose it in 32, 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 478 sessions in evals 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 evals run can be replayed on the board.
If you work on Arize Phoenix: 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 Arize Phoenix?
Yes. Arize Phoenix was chosen in 32 of the 478 judged sessions in evals, a 7% install share, ranking third in its category.
Does Claude Code recommend Arize Phoenix?
In 7 of the 120 sessions in evals run with Claude Code, which is 6%.
Do different coding agents treat Arize Phoenix differently?
Yes, and by a wide margin. Codex chose it in 11% of its runs and Muse Code in 0%.
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 Arize Phoenix considered but not chosen?
It was raised as a candidate in 131 sessions without being chosen, and chosen in 32. That is a 20% 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.