Benchmark Almanac

CMI-Bench

gemini-3.1-flash 41.8 tops its leaderboard Gemini 3.1 Flash-Lite 37.9 leads the models we track
Not in model cards

Models listen to music and answer requests such as naming an instrument, marking beats, or transcribing lyrics. Each task has its own scoring rule.

The benchmark paper has 53 Hugging Face upvotes. The authors evaluate 11 audio language models and publish the test audio and scoring code.

Released
June 2025
Kind
Audio understanding and voice
Level
University
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Subject
Arts
Language
English
In and out
Text, Audio → Text
Interaction
Single turn
Items from
Another benchmark
Refresh
One fixed set
Run by
Anyone
In views
Audio and speech
Strongest model tested
Gemini 3.1 Flash-Lite #96 on the capability index
Last model added
March 2026
Scores refreshed
November 20th, 2026
3
Entries on MuTheoryEval’s leaderboard, 1 model we track

gemini-3.1-flash tops the leaderboard with 41.8 Gemini 3.1 Flash-Lite leads with 37.9

Best score per entry from MuTheoryEval’s leaderboard, read on November 20th, 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 3.1 Flash-Lite, at 37.9. This view shows only models we track, each at its best.

  1. 1 gemini-3.1-flash lite, temperature=0, max_output_tokens=4096, thinking={"level":"default"}, answer_format=MULTI_SELECT 41.8
  2. 2 Gemini 3.1 Flash-Lite lite, temperature=0, max_output_tokens=4096, thinking={"level":"default"}, answer_format=MULTI_SELECT 37.9
  3. 3 gemini-3.1-flash (minimal) lite, temperature=0, max_output_tokens=4096, thinking={"level":"minimal"}, answer_format=MULTI_SELECT 37.1

Source: MuTheoryEval’s leaderboard

#EntryScoreSettingReleasedRun onSource
1 gemini-3.1-flash 41.8 lite, temperature=0, max_output_tokens=4096, thinking={"level":"default"}, answer_format=MULTI_SELECT Apr 2026 Official leaderboard
2 Gemini 3.1 Flash-Lite 37.9 lite, temperature=0, max_output_tokens=4096, thinking={"level":"default"}, answer_format=MULTI_SELECT Mar 2026 Apr 2026 Official leaderboard
3 gemini-3.1-flash (minimal) 37.1 lite, temperature=0, max_output_tokens=4096, thinking={"level":"minimal"}, answer_format=MULTI_SELECT Apr 2026 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 Gemini 3.1 Flash-Lite 37.9 lite, temperature=0, max_output_tokens=4096, thinking={"level":"default"}, answer_format=MULTI_SELECT Mar 2026 Apr 2026 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 MuTheoryEval’s leaderboard
37.9 2026 37.9 2026

Source: MuTheoryEval’s 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

7 of 8 checks pass, 1 warning. Look first at the Newest score check. The newest came 8 months ago, in March 2026, for Gemini 3.1 Flash-Lite. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 1 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (3 entries).
  • Note Models scored Only 1 model we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 8 months ago, in March 2026, for Gemini 3.1 Flash-Lite.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.1 Flash-Lite (#96).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on MuTheoryEval: 1 row for models we track and 2 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