Benchmark Almanac

Omnilingua-AudioMC

Muse Spark 1.2 64% leads since November 2026
Not in model cards

Models follow requirements across spoken conversations in ten languages. The authors compare each model's spoken and text responses against the same 1,404 rubric items.

Released
November 2026
Kind
Speech recognition and translation
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Audio, Text → Audio, Text
Interaction
Multi-turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 3.8 Flash #24 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
10
Entries in its paper, 4 models we track

Muse Spark 1.2 leads with 64% among models released by September 2026 Muse Spark 1.2 leads with 64% among models released by September 2026

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. 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 Muse Spark 1.2 Reference Transcript 64%
  2. 2 Gemini 3.1 Pro Reference Transcript 63%
  3. 3 Qwen3.8 Omni Reference Transcript 60.9%
  4. 4 Gemini 3.8 Flash Reference Transcript 57%
  5. 5 Doubao Seed Lite Reference Transcript 48.4%
  6. 6 Qwen3.5 Omni Reference Transcript 40.1%
  7. 7 Voxtral Small Reference Transcript 30.4%
  8. 8 Qwen3-Omni Reference Transcript 29%
  9. 9 Voxtral-Mini Reference Transcript 20.1%
  10. 10 Qwen2.5-Omni Reference Transcript 13.8%

Source: The Omnilingua-AudioMC paper

#EntryScoreSettingDateSource
1 Muse Spark 1.2 64% Reference Transcript Aug 2026 Official leaderboard
2 Gemini 3.1 Pro 63% Reference Transcript Feb 2026 Official leaderboard
3 Qwen3.8 Omni 60.9% Reference Transcript Official leaderboard
4 Gemini 3.8 Flash 57% Reference Transcript Sep 2026 Official leaderboard
5 Doubao Seed Lite 48.4% Reference Transcript Official leaderboard
6 Qwen3.5 Omni 40.1% Reference Transcript Official leaderboard
7 Voxtral Small 30.4% Reference Transcript Jul 2025 Official leaderboard
8 Qwen3-Omni 29% Reference Transcript Official leaderboard
9 Voxtral-Mini 20.1% Reference Transcript Official leaderboard
10 Qwen2.5-Omni 13.8% Reference Transcript Official leaderboard
#ModelScoreSettingDateSource
1 Muse Spark 1.2 64% Reference Transcript Aug 2026 Official leaderboard
2 Gemini 3.1 Pro 63% Reference Transcript Feb 2026 Official leaderboard
3 Gemini 3.8 Flash 57% Reference Transcript Sep 2026 Official leaderboard
4 Voxtral Small 30.4% Reference Transcript Jul 2025 Official leaderboard

Rankings by part

We read 10 parts from the benchmark’s paper, each ranking models on one part of the test, such as English, Chinese and Japanese. Gemini 3.1 Pro leads 5 of them, and Muse Spark 1.2 leads the other 5. Arrows show how far a model moved from its overall rank among the models we track.

English

4 models we track have a score here.

  1. 1 Gemini 3.1 Pro ↑1 66.9%
  2. 2 Muse Spark 1.2 ↓1 64.7%
  3. 3 Gemini 3.8 Flash 59.3%
  4. 4 Voxtral Small 29%

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Muse Spark 1.2
Direct Audio Current Tra… History Tra… Full Transc… Reference T…
64%
Gemini 3.1 Pro
Direct Audio History Tra…
63%
Gemini 3.8 Flash
Direct Audio Full Transc…
57%
Voxtral Small
Direct Audio
30.4%

Source: The Omnilingua-AudioMC paper

ModelEffortHarnessScoreCost per taskOutput tokens
Muse Spark 1.2 Reference Transcript 64%
Gemini 3.1 Pro Reference Transcript 63%
Gemini 3.1 Pro Full Transcript 60.8%
Gemini 3.1 Pro History Transcript 60.1%
Muse Spark 1.2 Full Transcript 58.6%
Gemini 3.1 Pro Current Transcript 57.1%
Gemini 3.8 Flash Reference Transcript 57%
Gemini 3.1 Pro Direct Audio 54.4%
Gemini 3.8 Flash Full Transcript 54.4%
Gemini 3.8 Flash History Transcript 53.2%
Muse Spark 1.2 History Transcript 52.5%
Gemini 3.8 Flash Current Transcript 50.8%
Gemini 3.8 Flash Direct Audio 50%
Muse Spark 1.2 Current Transcript 38.6%
Muse Spark 1.2 Direct Audio 32.4%
Voxtral Small Reference Transcript 30.4%
Voxtral Small Direct Audio 28.1%

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Muse Spark 1.2
Direct Audio Current Tra… History Tra… Full Transc… Reference T…
64%
Gemini 3.1 Pro
Direct Audio History Tra…
63%
Gemini 3.8 Flash
Direct Audio Full Transc…
57%
Voxtral Small
Direct Audio
30.4%

Source: The Omnilingua-AudioMC paper

ModelEffortHarnessScoreCost per taskOutput tokens
Muse Spark 1.2 Reference Transcript 64%
Gemini 3.1 Pro Reference Transcript 63%
Gemini 3.1 Pro Full Transcript 60.8%
Gemini 3.1 Pro History Transcript 60.1%
Muse Spark 1.2 Full Transcript 58.6%
Gemini 3.1 Pro Current Transcript 57.1%
Gemini 3.8 Flash Reference Transcript 57%
Gemini 3.1 Pro Direct Audio 54.4%
Gemini 3.8 Flash Full Transcript 54.4%
Gemini 3.8 Flash History Transcript 53.2%
Muse Spark 1.2 History Transcript 52.5%
Gemini 3.8 Flash Current Transcript 50.8%
Gemini 3.8 Flash Direct Audio 50%
Muse Spark 1.2 Current Transcript 38.6%
Muse Spark 1.2 Direct Audio 32.4%
Voxtral Small Reference Transcript 30.4%
Voxtral Small Direct Audio 28.1%

Number one over time

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

Meta
Benchmark paper 1 leader

Source: The Omnilingua-AudioMC paper

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
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Muse Spark 1.2 Gemini 3.1 Pro Voxtral Small 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Muse Spark 1.2 Voxtral Small

Source: Official Omnilingua-AudioMC 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

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (10 entries).
  • Pass Models scored 4 models we track, plus 6 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for Gemini 3.8 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.8 Flash (#24).
  • Pass Headroom The best is 64%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 17 rows for models we track and 6 other entries at the last read.
  • Pass Last read November 20th, 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