Benchmark Almanac

StudyArena Audio

Kokoro 82M 1,229 tops its leaderboard MiniMax Speech 2.8 Turbo 1,216 leads the models we track
Not in model cards

Models produce audio responses to user prompts. Anonymous human comparisons determine the published rating in this separate vote pool.

Only entries with observed ballots count as results. Topic filters do not form separate benchmark records.

Made by
StudyArena
Kind
Audio understanding and voice
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Last model added
January 2026
Scores refreshed
November 12th, 2026
10
Entries on its own leaderboard, 2 models we track

Kokoro 82M tops the leaderboard with 1,229

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is MiniMax Speech 2.8 Turbo, at 1,216.

Bars start at 1,120.

  1. 1,229 Kokoro 82M
  2. 1,216 Sesame CSM 1B
  3. 1,216 MiniMax Speech 2.8 Turbo
  4. 1,215 Eleven v3
  5. 1,215 MiniMax Speech 2.8 HD
  6. 1,201 Grok Voice TTS 1.0
  7. 1,200 Qwen Audio 3.0 TTS Flash
  8. 1,184 Aura-2
  9. 1,184 Voxtral Mini TTS
  10. 1,140 Chatterbox HD

Bars start at 1,120.

  1. 1 Kokoro 82M 1,229
  2. 2 Sesame CSM 1B 1,216
  3. 3 MiniMax Speech 2.8 Turbo 1,216
  4. 4 Eleven v3 1,215
  5. 5 MiniMax Speech 2.8 HD 1,215
  6. 6 Grok Voice TTS 1.0 1,201
  7. 7 Qwen Audio 3.0 TTS Flash 1,200
  8. 8 Aura-2 1,184
  9. 9 Voxtral Mini TTS 1,184
  10. 10 Chatterbox HD 1,140

Source: Official StudyArena Audio leaderboard

#EntryScoreSettingDateSource
1 Kokoro 82M 1,229 Official leaderboard
2 Sesame CSM 1B 1,216 Official leaderboard
3 MiniMax Speech 2.8 Turbo 1,216 Jan 2026 Official leaderboard
4 Eleven v3 1,215 Official leaderboard
5 MiniMax Speech 2.8 HD 1,215 Jan 2026 Official leaderboard
6 Grok Voice TTS 1.0 1,201 Official leaderboard
7 Qwen Audio 3.0 TTS Flash 1,200 Official leaderboard
8 Aura-2 1,184 Official leaderboard
9 Voxtral Mini TTS 1,184 Official leaderboard
10 Chatterbox HD 1,140 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
1,215 1,215 1,215 1,216 1,216 1,216 2026 1,215 1,215 1,215 1,216 1,216 1,216 2026

Source: Official StudyArena Audio leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Audio and video

All 232 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%