Artificial Analysis Ranks SpeechifyAI Simba 3.2 First for TTS Quality and Price

Artificial AnalysisArtificial Analysis

Artificial Analysis ranked SpeechifyAI’s Simba 3.2 as the top text-to-speech model on its Speech Arena leaderboard following blind human preference testing. The model achieved a 1,233 Elo score, surpassing previous leaders from Google and Cartesia while maintaining the lowest price point among the top five models.

Artificial Analysis ranked SpeechifyAI’s Simba 3.2 first on its Speech Arena leaderboard. The model achieved an Elo score of 1,233 based on human preference wins. This follows the launch of the Artificial Analysis Controlled Voice Arena, which standardizes evaluation by cloning models onto identical voices to isolate quality.
Elo score
1,233
Pricing
$10.00 per 1M characters
Speed
29.2 characters per second
Availability
SpeechifyAI API

The finding establishes a new Pareto frontier for text-to-speech by combining top-tier naturalness with aggressive pricing. Simba 3.2 outranked Fish Audio’s S2.1 Pro benchmark in quality while costing significantly less. At $10 per million characters, it is nearly 4x cheaper than Cartesia’s Sonic 3.5, which previously held a top-three position.

Simba 3.2 generates 29.2 characters per second, trailing specialized real-time models in speed. This shift suggests high-fidelity voice is becoming a commodity, pressuring established providers to justify their premiums. You can access Simba 3.2 via the SpeechifyAI API at $10 per million characters.

Artificial Analysis
Artificial Analysis
@ArtificialAnlys
X

SpeechifyAI’s Simba 3.2 takes the #1 spot on the Artificial Analysis Speech Arena Leaderboard, surpassing Sonic 3.5, Inworld Realtime TTS 1.5 Max and Google’s Gemini 3.1 Flash TTS Simba 3.2 is the latest TTS model from @SpeechifyAI. Key takeaways: ➤ Quality: Simba 3.2 has an Elo score of 1,233 (+17/-17) based on 1,258 arena appearances, placing it ahead of Gemini 3.1 Flash TTS at 1,214 and Sonic 3.5 at 1,210. ➤ Pricing: At $10/1M characters, Simba 3.2 is the lowest-priced model among the current Speech Arena leaders, compared to Gemini 3.1 Flash TTS ($18.3/1M characters), Inworld Realtime TTS 1.5 Max ($26/1M characters), and Sonic 3.5 ($39/1M characters). ➤ Speed: 29.2 characters per second, compared to 89 characters per second for Inworld Realtime TTS 1.5 Max and 24.5 characters per second for Gemini 3.1 Flash TTS. See more details and listen to samples below 🧵

8retweets36likes
View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Share this update