The devtools AI chooses.
We give LLMs the same app-building prompts, then rank the tools they recommend.
Ranked by recommendations from
Loading model recommendations…
How we measure
We give LLMs the same app-building prompts, then rank the tools they recommend.
Ranked by recommendations from
Loading model recommendations…
Loading rankings...
| # | Tool | Support Rate |
|---|---|---|
| 1 |  LALangSmith | 52.7%(2417/4589) |
| 2 |  LALangfuse | 29.7%(1365/4589) |
| 3 | ARArize AI | 6.9%(315/4589) |
| 4 |  DADatadog | 1.2%(53/4589) |
| 5 |  WEWeights & Biases | 1.2%(53/4589) |
| 6 |  ARArize Phoenix | 1.1%(50/4589) |
| 7 |  HEHelicone | 0.9%(42/4589) |
| 8 |  LALangChain | 0.7%(30/4589) |
| 9 | PRPromptLayer | 0.5%(25/4589) |
| 10 |  BRBraintrust | 0.3%(16/4589) |
| 11 |  OPOpenTelemetry | 0.3%(16/4589) |
| 12 | VEVellum | 0.2%(9/4589) |
| 13 | HUHumanloop | 0.2%(7/4589) |
| 14 | LLLlamaIndex | 0.2%(7/4589) |
| 15 | INIntercom | 0.1%(5/4589) |
| 16 | ZEZendesk | 0.1%(5/4589) |
| 17 | HOHoneycomb | 0.1%(4/4589) |
| 18 | FIFireflies.ai | 0.1%(3/4589) |
| 19 | NENew Relic | 0.1%(3/4589) |
| 20 |  POPostHog | 0.0%(2/4589) |
| 21 | HOHoneyHive | 0.0%(1/4589) |
| 22 |  PAPatronus AI | 0.0%(1/4589) |
| 23 | POPortkey | 0.0%(1/4589) |