Murf AI models, access and benchmark guide
Murf AI model and access guide with a practical US-English evaluation brief and links to publisher materials.

👁 111 reads
A voice-selection page and a real-time API solve different parts of a product. For Murf’s Falcon, use the synthesis endpoint’s character tariff and measure the first audible response to a short line, separate from the duration of the complete file.
Choosing the first job
Read a fictional departure gate twice with a corrected letter; record the latency and which gate the listener heard.
Models in the November 17 snapshot
| Model and track | Placement | Reported score |
|---|---|---|
| Falcon 2.1 (AA provider voice) | 33 | 1153 Elo |
Release record. Publisher documentation identifies the product family. The ranked configurations include Falcon 2.1.
Google · Your sourcesQuantslant higher in your Google - one clickAdd to preferred sources →Sources
- https://murf.ai/api (accessed November 17, 2026)
- https://murf.ai/api/docs/introduction/overview (accessed November 17, 2026)
- https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice (accessed November 17, 2026)
More on this from QUANTSLANT
benchmarks·Mureka models, access and benchmark guideMureka model and access guide with a practical US-English evaluation brief and links to publisher materials.