Benchmark Almanac

SuperCLUE-TTS

ULingo-TTS-1.0-Preview 71.4 tops its leaderboard Gemini 3.8 Flash TTS 69.7 leads the models we track
Not in model cards

Models read Chinese text aloud. The benchmark rates basic speech synthesis and delivery in situations such as public announcements.

SuperCLUE publishes scores for 17 models in its 2026 results workbook.

Made by
SuperCLUE
Released
December 2024
Kind
Audio understanding and voice
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Measures
Capability
Language
Chinese
In and out
Text → Audio
Interaction
Single turn
Refresh
One fixed set
Run by
A host
In views
Audio and speech
Last model added
September 2026
Scores refreshed
November 21st, 2026
17
Entries on its own leaderboard, 4 models we track

ULingo-TTS-1.0-Preview tops the leaderboard with 71.4 Gemini 3.8 Flash TTS leads with 69.7 among models released by September 2026

Best score per entry from the benchmark's own leaderboard, read on November 21st, 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 Gemini 3.8 Flash TTS, at 69.7. This view shows only models we track, each at its best.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 ULingo-TTS-1.0-Preview 71.4
  2. 2 Seed-TTS-2.0-Standard 70.8
  3. 3 Gemini 3.8 Flash TTS 69.7
  4. 4 Sonic 3.6 69.6
  5. 5 Qwen-Audio-3.1-TTS-Flash 68.4
  6. 6 Qwen-Audio-3.0-TTS-Plus 68.3
  7. 7 Gemini 3.1 Flash TTS 66.6
  8. 8 StepAudio-3-TTS 65.4
  9. 9 MiniMax Speech 2.8 HD 64.4
  10. 10 MiMo-V2.5-TTS 63.4
  11. 11 Sonic 3.5 61.8
  12. 12 超拟人语音合成 59.9
  13. 13 StepAudio 2.5 TTS 59.4
  14. 14 VoxCPM2 59.3
  15. 15 Llama Realtime TTS-2 59.2
  16. 16 Qwen3-TTS-Flash 57.5
  17. 17 MOSS-TTS-Local-Transformer-v1.5 55.0

Source: Official SuperCLUE-TTS leaderboard

#EntryScoreSettingReleasedRun onSource
1 ULingo-TTS-1.0-Preview closed weights 71.4 Oct 2026 Official leaderboard
2 Seed-TTS-2.0-Standard closed weights 70.8 Jul 2026 Official leaderboard
3 Gemini 3.8 Flash TTS 69.7 Sep 2026 Oct 2026 Official leaderboard
4 Sonic 3.6 closed weights 69.6 Oct 2026 Official leaderboard
5 Qwen-Audio-3.1-TTS-Flash closed weights 68.4 Oct 2026 Official leaderboard
6 Qwen-Audio-3.0-TTS-Plus closed weights 68.3 Jul 2026 Official leaderboard
7 Gemini 3.1 Flash TTS 66.6 Apr 2026 Jul 2026 Official leaderboard
8 StepAudio-3-TTS closed weights 65.4 Oct 2026 Official leaderboard
9 MiniMax Speech 2.8 HD 64.4 Jan 2026 Jul 2026 Official leaderboard
10 MiMo-V2.5-TTS closed weights 63.4 Jul 2026 Official leaderboard
11 Sonic 3.5 closed weights 61.8 Jul 2026 Official leaderboard
12 超拟人语音合成 closed weights 59.9 Jul 2026 Official leaderboard
13 StepAudio 2.5 TTS closed weights 59.4 Jul 2026 Official leaderboard
14 VoxCPM2 open weights 59.3 Jul 2026 Official leaderboard
15 Llama Realtime TTS-2 closed weights 59.2 Jul 2026 Official leaderboard
16 Qwen3-TTS-Flash 57.5 Sep 2025 Jul 2026 Official leaderboard
17 MOSS-TTS-Local-Transformer-v1.5 open weights 55.0 Jul 2026 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 Gemini 3.8 Flash TTS 69.7 Sep 2026 Oct 2026 Official leaderboard
2 Gemini 3.1 Flash TTS 66.6 Apr 2026 Jul 2026 Official leaderboard
3 MiniMax Speech 2.8 HD 64.4 Jan 2026 Jul 2026 Official leaderboard
4 Qwen3-TTS-Flash 57.5 Sep 2025 Jul 2026 Official leaderboard

Rankings by part

We read 15 parts from the benchmark’s own leaderboard in 2 groups: 基础能力 and 场景应用. Each ranks models on one part of the test. Gemini 3.8 Flash TTS leads 10 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

基础能力得分

4 models we track have a score here.

  1. 1 Gemini 3.8 Flash TTS 72.9
  2. 2 Gemini 3.1 Flash TTS 70.7
  3. 3 MiniMax Speech 2.8 HD 70.1
  4. 4 Qwen3-TTS-Flash 60.5

4 models have held number one on the official leaderboard since its release in December 2024

Lab claims count from the day the card came out, independent runs from the model’s release.

Qwen MiniMax Google DeepMind
Own leaderboard Not tested since Sep 2026

Source: Official SuperCLUE-TTS leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
57.5 60.0 62.5 65.0 67.5 70.0 2026 57.5 60.0 62.5 65.0 67.5 70.0 2026

Source: Official SuperCLUE-TTS leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Audio and video

All 695 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%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (17 entries).
  • Pass Models scored 4 models we track, plus 13 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for Gemini 3.8 Flash TTS.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 4 rows for models we track and 13 other entries at the last read.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check