Benchmark Almanac

MusiXQA

Phi-3-MusiX 81.4% tops its leaderboard Phi-3-vision-128k-instruct 12.8% leads the models we track
Not in model cards

Models read printed music and answer questions about notes, chords, text, and page layout. The authors measure answer accuracy and string similarity.

The Legato 2 team ran Gemini 3 Flash and GPT-5 mini on the notation recognition split. They used GPT-5 to grade answers.

Released
June 2025
Kind
Documents, charts and text in images
Level
School
Scored
Graded by a model
Answers
Short answer
Tasks
All public
Measures
Capability
Subject
Arts
Language
English
In and out
Text, Images, Documents → Text
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
In views
Vision
Last model added
May 2024
Scores refreshed
November 20th, 2026
3
Entries in its paper, 1 model we track

Phi-3-MusiX tops the leaderboard with 81.4% Phi-3-vision-128k-instruct leads with 12.8%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the mean GPT-graded accuracy across four sheet music tasks. 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 Phi-3-vision-128k-instruct, at 12.8%. This view shows only models we track, each at its best.

  1. 1 Phi-3-MusiX kern+ 81.4%
  2. 2 GPT-4o (snapshot unspecified) 39%
  3. 3 Phi-3-vision-128k-instruct Zero-shot 12.8%

Source: The MusiXQA paper

Source: The MusiXQA paper

#EntryScoreSettingDateSource
1 Phi-3-MusiX 81.4% kern+ Official leaderboard
2 GPT-4o (snapshot unspecified) 39% RAG + OMR Official leaderboard
3 Phi-3-vision-128k-instruct 12.8% Zero-shot May 2024 Official leaderboard
#ModelScoreSettingDateSource
1 Phi-3-vision-128k-instruct 12.8% Zero-shot May 2024 Official leaderboard

Scores over time

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

In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Phi-3-vision-128k-instruct 0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Phi-3-vision-128k-instruct

Source: Official MusiXQA 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 Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in May 2024, for Phi-3-vision-128k-instruct. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (3 entries).
  • Note Models scored Only 1 model we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 2 years ago, in May 2024, for Phi-3-vision-128k-instruct.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 12.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 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