TTS API comparison 2026
Gradium TTS Beta vs Inworld Realtime TTS-2 Flash Gradium vs Inworld · text-to-speech for voice agents
Gradium TTS Beta ranks 1st of 112 with 76.8/100; Inworld Realtime TTS-2 Flash ranks 5th with 68.4/100: a gap of 8.4 points on the same public rules, built on Coval, Speko and Artificial Analysis.
Data checked on September 30, 2026 · Coval 30-day window ending October 1, 2026 · updated after each benchmark capture
Side-by-side benchmark
| Measure | Gradium TTS Beta | Inworld Realtime TTS-2 Flash |
|---|---|---|
| Overall score | 76.8/100 | 68.4/100 |
| Rank | 1st of 112 | 5th of 112 |
| Real-time speed /60 | 58.7 | 43.5 |
| Voice quality /20 | 7.8 | 13.1 |
| Accuracy /15 | 9.8 | 7.8 |
| Price /5 | 0.5 | 4 |
| P50 / median time-to-first-audio Coval, 30-day window | 47.2 ms | 69.4 ms |
| P99 time-to-first-audio (slowest turns) Coval, 30-day window | 108.8 ms | 175.2 ms |
| Latency consistency (standard deviation) Coval, 30-day window | 16.6 ms | 373.3 ms |
| P50 time-to-first-audio, US East Speko | 100 ms | 82 ms |
| Accuracy: word error rate (WER) Coval, 30-day window | 4.5% | 5.4% |
| Pronunciation robustness (0 to 1) Speko | 0.3 | 0.6 |
| Speech Arena Elo (blind listener preference) Artificial Analysis | not measured | 1209 |
| Naturalness (Elo) Speko | 1579 | 1501 |
| Price per 1M characters Artificial Analysis | not measured | 10.4 USD |
Choose Gradium TTS Beta if…
- → Ranks 1st of 112 in the overall benchmark, against 5th for Inworld Realtime TTS-2 Flash
- → Better P50 / median time-to-first-audio: 47.2 ms vs 69.4 ms (Coval, 30-day window)
- → Better P99 time-to-first-audio (slowest turns): 108.8 ms vs 175.2 ms (Coval, 30-day window)
- → Better latency consistency (standard deviation): 16.6 ms vs 373.3 ms (Coval, 30-day window)
- → Better accuracy: word error rate (WER): 4.5% vs 5.4% (Coval, 30-day window)
- → Better naturalness (Elo): 1579 vs 1501 (Speko)
Choose Inworld Realtime TTS-2 Flash if…
- → Better P50 time-to-first-audio, US East: 82 ms vs 100 ms (Speko)
- → Better pronunciation robustness (0 to 1): 0.6 vs 0.3 (Speko)
Frequently asked questions
- What is the difference between Gradium TTS Beta and Inworld Realtime TTS-2 Flash?
- In the 2026 TTS API benchmark, Gradium TTS Beta ranks 1st of 112 with 76.8/100 and Inworld Realtime TTS-2 Flash ranks 5th with 68.4/100, a gap of 8.4 points on the same rules.
- Which is faster for real-time voice agents, Gradium TTS Beta or Inworld Realtime TTS-2 Flash?
- On Coval's 30-day window ending October 1, 2026, the P50 time-to-first-audio is 47.2 ms for Gradium TTS Beta and 69.4 ms for Inworld Realtime TTS-2 Flash; the P99 is 108.8 ms and 175.2 ms.
- On which criteria does Inworld Realtime TTS-2 Flash beat Gradium TTS Beta?
- Inworld Realtime TTS-2 Flash scores higher on Voice quality (13.1 vs 7.8 out of 20), Price (4 vs 0.5 out of 5).
- How are these two text-to-speech APIs scored?
- Both are scored by the same public rules: 4 criteria, 100 points (Real-time speed 60, Voice quality 20, Accuracy 15, Price 5), computed from three independent benchmarks: Coval, Speko and Artificial Analysis. A measurement a benchmark does not publish counts zero, and the model page says why.
- Can these numbers be checked?
- Yes. Each model page lists every measurement with its source and date, and links to the benchmark page it comes from. The rules are on the methodology page.