PianoBench
Models answer questions about piano performances from recordings or symbolic transcriptions. The authors compare accuracy with and without access to the music.
- Released
- November 2026
- Kind
- Audio understanding and voice
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Arts
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Audio and speech
- Strongest model tested
- GPT-6 Astra #4 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 20th, 2026
GPT-6 Astra leads with 93.2% among models released by September 2026 GPT-6 Astra leads with 93.2% 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:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- Claude Opus 5.5
- and 13 more
- 93.2 GPT-6 Astra
- 90.0
Gemini 3.7 Flash
- 89.2
Gemini 3.1 Pro
- 88.8
Gemini 3.8 Flash
- 88.0 GPT-5.6 Sol
- 86.8
Muse Spark 1.2
- 85.2 GPT-5.6 Terra
- 70.0
Qwen3.5-9B
- 68.8
Inkling-Small
- 66.4
Inkling
- 63.2
Qwen3.5-Omni-Plus
- 58.4
Gemma 4 12B
- 50.8
Qwen3-Omni-30B-A3B
- 48.8 Nemotron Audex 30B-A3B
- 48.4
Qwen3-Omni-30B-A3B-Thinking
- 48.4
Nemotron 3 Nano Omni 30B-A3B
- 45.6 Music Flamingo
- 38.8
Audio Flamingo 3
- 1 GPT-6 Astra Zero-shot, MIDI-text, high 93.2%
- 2
Gemini 3.7 Flash Zero-shot, MIDI-text, medium 90%
- 3
Gemini 3.1 Pro Zero-shot, MIDI-text, medium 89.2%
- 4
Gemini 3.8 Flash Zero-shot, MIDI-text, medium 88.8%
- 5 GPT-5.6 Sol Zero-shot, MIDI-text, xhigh 88%
- 6
Muse Spark 1.2 Zero-shot, MIDI-text, medium 86.8%
- 7 GPT-5.6 Terra Zero-shot, MIDI-text, high 85.2%
- 8
Qwen3.5-9B Zero-shot, MIDI-text, thinking 70%
- 9
Inkling-Small Zero-shot, MIDI-text, medium 68.8%
- 10
Inkling Zero-shot, MIDI-text, medium 66.4%
- 11
Qwen3.5-Omni-Plus Zero-shot, Piano stem 63.2%
- 12
Gemma 4 12B Zero-shot, MIDI-text, thinking 58.4%
- 13
Qwen3-Omni-30B-A3B Zero-shot, Piano synth, non-thinking 50.8%
- 14 Nemotron Audex 30B-A3B Zero-shot, MIDI-text 48.8%
- 15
Qwen3-Omni-30B-A3B-Thinking Zero-shot, Piano synth, thinking 48.4%
- 16
Nemotron 3 Nano Omni 30B-A3B Zero-shot, Piano stem 48.4%
- 17 Music Flamingo Zero-shot, Piano synth 45.6%
- 18
Audio Flamingo 3 Zero-shot, Piano synth 38.8%
Source: The PianoBench paper
- 93.2 GPT-6 Astra
- 90.0
Gemini 3.7 Flash
- 89.2
Gemini 3.1 Pro
- 88.8
Gemini 3.8 Flash
- 88.0 GPT-5.6 Sol
- 86.8
Muse Spark 1.2
- 85.2 GPT-5.6 Terra
- 70.0
Qwen3.5-9B
- 68.8
Inkling-Small
- 66.4
Inkling
- 63.2
Qwen3.5-Omni-Plus
- 58.4
Gemma 4 12B
- 50.8
Qwen3-Omni-30B-A3B
- 48.4
Qwen3-Omni-30B-A3B-Thinking
- 48.4
Nemotron 3 Nano Omni 30B-A3B
- 38.8
Audio Flamingo 3
- 1 GPT-6 Astra Zero-shot, MIDI-text, high 93.2%
- 2
Gemini 3.7 Flash Zero-shot, MIDI-text, medium 90%
- 3
Gemini 3.1 Pro Zero-shot, MIDI-text, medium 89.2%
- 4
Gemini 3.8 Flash Zero-shot, MIDI-text, medium 88.8%
- 5 GPT-5.6 Sol Zero-shot, MIDI-text, xhigh 88%
- 6
Muse Spark 1.2 Zero-shot, MIDI-text, medium 86.8%
- 7 GPT-5.6 Terra Zero-shot, MIDI-text, high 85.2%
- 8
Qwen3.5-9B Zero-shot, MIDI-text, thinking 70%
- 9
Inkling-Small Zero-shot, MIDI-text, medium 68.8%
- 10
Inkling Zero-shot, MIDI-text, medium 66.4%
- 11
Qwen3.5-Omni-Plus Zero-shot, Piano stem 63.2%
- 12
Gemma 4 12B Zero-shot, MIDI-text, thinking 58.4%
- 13
Qwen3-Omni-30B-A3B Zero-shot, Piano synth, non-thinking 50.8%
- 14
Qwen3-Omni-30B-A3B-Thinking Zero-shot, Piano synth, thinking 48.4%
- 15
Nemotron 3 Nano Omni 30B-A3B Zero-shot, Piano stem 48.4%
- 16
Audio Flamingo 3 Zero-shot, Piano synth 38.8%
Source: The PianoBench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 93.2% | Zero-shot, MIDI-text, high | Sep 2026 | Official leaderboard |
| 2 | | 90% | Zero-shot, MIDI-text, medium | Aug 2026 | Official leaderboard |
| 3 | | 89.2% | Zero-shot, MIDI-text, medium | Feb 2026 | Official leaderboard |
| 4 | | 88.8% | Zero-shot, MIDI-text, medium | Sep 2026 | Official leaderboard |
| 5 | GPT-5.6 Sol | 88% | Zero-shot, MIDI-text, xhigh | Jul 2026 | Official leaderboard |
| 6 | | 86.8% | Zero-shot, MIDI-text, medium | Aug 2026 | Official leaderboard |
| 7 | GPT-5.6 Terra | 85.2% | Zero-shot, MIDI-text, high | Jul 2026 | Official leaderboard |
| 8 | | 70% | Zero-shot, MIDI-text, thinking | Feb 2026 | Official leaderboard |
| 9 | | 68.8% | Zero-shot, MIDI-text, medium | Jul 2026 | Official leaderboard |
| 10 | | 66.4% | Zero-shot, MIDI-text, medium | Jul 2026 | Official leaderboard |
| 11 | | 63.2% | Zero-shot, Piano stem | Mar 2026 | Official leaderboard |
| 12 | | 58.4% | Zero-shot, MIDI-text, thinking | Jun 2026 | Official leaderboard |
| 13 | | 50.8% | Zero-shot, Piano synth, non-thinking | Sep 2025 | Official leaderboard |
| 14 | Nemotron Audex 30B-A3B | 48.8% | Zero-shot, MIDI-text | Official leaderboard | |
| 15 | | 48.4% | Zero-shot, Piano synth, thinking | Sep 2025 | Official leaderboard |
| 16 | | 48.4% | Zero-shot, Piano stem | Apr 2026 | Official leaderboard |
| 17 | Music Flamingo | 45.6% | Zero-shot, Piano synth | Official leaderboard | |
| 18 | | 38.8% | Zero-shot, Piano synth | Jul 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 93.2% | Zero-shot, MIDI-text, high | Sep 2026 | Official leaderboard |
| 2 | | 90% | Zero-shot, MIDI-text, medium | Aug 2026 | Official leaderboard |
| 3 | | 89.2% | Zero-shot, MIDI-text, medium | Feb 2026 | Official leaderboard |
| 4 | | 88.8% | Zero-shot, MIDI-text, medium | Sep 2026 | Official leaderboard |
| 5 | GPT-5.6 Sol | 88% | Zero-shot, MIDI-text, xhigh | Jul 2026 | Official leaderboard |
| 6 | | 86.8% | Zero-shot, MIDI-text, medium | Aug 2026 | Official leaderboard |
| 7 | GPT-5.6 Terra | 85.2% | Zero-shot, MIDI-text, high | Jul 2026 | Official leaderboard |
| 8 | | 70% | Zero-shot, MIDI-text, thinking | Feb 2026 | Official leaderboard |
| 9 | | 68.8% | Zero-shot, MIDI-text, medium | Jul 2026 | Official leaderboard |
| 10 | | 66.4% | Zero-shot, MIDI-text, medium | Jul 2026 | Official leaderboard |
| 11 | | 63.2% | Zero-shot, Piano stem | Mar 2026 | Official leaderboard |
| 12 | | 58.4% | Zero-shot, MIDI-text, thinking | Jun 2026 | Official leaderboard |
| 13 | | 50.8% | Zero-shot, Piano synth, non-thinking | Sep 2025 | Official leaderboard |
| 14 | | 48.4% | Zero-shot, Piano synth, thinking | Sep 2025 | Official leaderboard |
| 15 | | 48.4% | Zero-shot, Piano stem | Apr 2026 | Official leaderboard |
| 16 | | 38.8% | Zero-shot, Piano synth | Jul 2025 | Official leaderboard |
More effort adds a median 28 points
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.
Source: The PianoBench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | Zero-shot, MIDI-text, high | 93.2% | |||
| Zero-shot, MIDI-text, medium | 90% | ||||
| Zero-shot, MIDI-text, medium | 89.2% | ||||
| Zero-shot, MIDI-text, medium | 88.8% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, xhigh | 88% | |||
| GPT-5.6 Sol | Zero-shot, MIDI-text, high | 87.2% | |||
| Zero-shot, MIDI-text, medium | 86.8% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, medium | 85.6% | |||
| GPT-5.6 Terra | Zero-shot, MIDI-text, high | 85.2% | |||
| Zero-shot, Piano stem, medium | 80.4% | ||||
| Zero-shot, Piano synth, medium | 78.4% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, none | 70.4% | |||
| Zero-shot, MIDI-text, thinking | 70% | ||||
| Zero-shot, MIDI-text, medium | 68.8% | ||||
| Zero-shot, Piano stem, medium | 66.8% | ||||
| Zero-shot, MIDI-text, medium | 66.4% | ||||
| Zero-shot, Piano synth, medium | 66.4% | ||||
| Zero-shot, Piano stem, medium | 64% | ||||
| Zero-shot, No music, medium | 63.6% | ||||
| Zero-shot, Piano synth, medium | 63.6% | ||||
| Zero-shot, Piano stem | 63.2% | ||||
| Zero-shot, Piano synth | 62.4% | ||||
| Zero-shot, No music, medium | 62% | ||||
| Zero-shot, No music, medium | 62% | ||||
| GPT-5.6 Sol | Zero-shot, No music, xhigh | 58.8% | |||
| Zero-shot, MIDI-text, thinking | 58.4% | ||||
| Zero-shot, MIDI-text | 57.2% | ||||
| Zero-shot, Piano synth, medium | 56.8% | ||||
| GPT-6 Astra | Zero-shot, No music, high | 56.8% | |||
| GPT-5.6 Sol | Zero-shot, No music, high | 56.4% | |||
| GPT-5.6 Sol | Zero-shot, No music, medium | 56% | |||
| Zero-shot, Piano stem, medium | 55.6% | ||||
| Zero-shot, Piano synth, medium | 52.8% | ||||
| GPT-5.6 Sol | Zero-shot, No music, none | 52.4% | |||
| Zero-shot, No music | 52% | ||||
| Zero-shot, Piano stem, non-thinking | 51.6% | ||||
| Zero-shot, No music, medium | 51.6% | ||||
| Zero-shot, Piano synth, non-thinking | 51.2% | ||||
| Zero-shot, Piano synth, non-thinking | 50.8% | ||||
| Zero-shot, Piano stem, medium | 50.4% | ||||
| Zero-shot, MIDI-text, non-thinking | 50% | ||||
| Zero-shot, Piano stem, non-thinking | 49.2% | ||||
| Zero-shot, No music, medium | 48.8% | ||||
| Zero-shot, Piano synth, thinking | 48.4% | ||||
| Zero-shot, Piano stem | 48.4% | ||||
| Zero-shot, MIDI-text, thinking | 47.6% | ||||
| Zero-shot, Piano synth, medium | 47.6% | ||||
| Zero-shot, MIDI-text, non-thinking | 47.2% | ||||
| Zero-shot, Piano synth | 47.2% | ||||
| GPT-5.6 Terra | Zero-shot, No music, high | 47.2% | |||
| Zero-shot, No music, medium | 47.2% | ||||
| Zero-shot, Piano stem, thinking | 46.8% | ||||
| Zero-shot, Piano stem, medium | 46.8% | ||||
| Zero-shot, No music, non-thinking | 46.4% | ||||
| Zero-shot, No music | 46.4% | ||||
| Zero-shot, Piano synth, thinking | 46.4% | ||||
| Zero-shot, MIDI-text | 45.6% | ||||
| Zero-shot, Piano stem, thinking | 45.2% | ||||
| Zero-shot, No music, non-thinking | 44.8% | ||||
| Zero-shot, No music, thinking | 44.8% | ||||
| Zero-shot, No music, thinking | 43.6% | ||||
| Zero-shot, No music, thinking | 42.4% | ||||
| Zero-shot, Piano synth | 38.8% | ||||
| Zero-shot, Piano stem | 37.2% |
More effort adds a median 28 points
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.
Source: The PianoBench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | Zero-shot, MIDI-text, high | 93.2% | |||
| Zero-shot, MIDI-text, medium | 90% | ||||
| Zero-shot, MIDI-text, medium | 89.2% | ||||
| Zero-shot, MIDI-text, medium | 88.8% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, xhigh | 88% | |||
| GPT-5.6 Sol | Zero-shot, MIDI-text, high | 87.2% | |||
| Zero-shot, MIDI-text, medium | 86.8% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, medium | 85.6% | |||
| GPT-5.6 Terra | Zero-shot, MIDI-text, high | 85.2% | |||
| Zero-shot, Piano stem, medium | 80.4% | ||||
| Zero-shot, Piano synth, medium | 78.4% | ||||
| GPT-5.6 Sol | Zero-shot, MIDI-text, none | 70.4% | |||
| Zero-shot, MIDI-text, thinking | 70% | ||||
| Zero-shot, MIDI-text, medium | 68.8% | ||||
| Zero-shot, Piano stem, medium | 66.8% | ||||
| Zero-shot, MIDI-text, medium | 66.4% | ||||
| Zero-shot, Piano synth, medium | 66.4% | ||||
| Zero-shot, Piano stem, medium | 64% | ||||
| Zero-shot, No music, medium | 63.6% | ||||
| Zero-shot, Piano synth, medium | 63.6% | ||||
| Zero-shot, Piano stem | 63.2% | ||||
| Zero-shot, Piano synth | 62.4% | ||||
| Zero-shot, No music, medium | 62% | ||||
| Zero-shot, No music, medium | 62% | ||||
| GPT-5.6 Sol | Zero-shot, No music, xhigh | 58.8% | |||
| Zero-shot, MIDI-text, thinking | 58.4% | ||||
| Zero-shot, MIDI-text | 57.2% | ||||
| Zero-shot, Piano synth, medium | 56.8% | ||||
| GPT-6 Astra | Zero-shot, No music, high | 56.8% | |||
| GPT-5.6 Sol | Zero-shot, No music, high | 56.4% | |||
| GPT-5.6 Sol | Zero-shot, No music, medium | 56% | |||
| Zero-shot, Piano stem, medium | 55.6% | ||||
| Zero-shot, Piano synth, medium | 52.8% | ||||
| GPT-5.6 Sol | Zero-shot, No music, none | 52.4% | |||
| Zero-shot, No music | 52% | ||||
| Zero-shot, Piano stem, non-thinking | 51.6% | ||||
| Zero-shot, No music, medium | 51.6% | ||||
| Zero-shot, Piano synth, non-thinking | 51.2% | ||||
| Zero-shot, Piano synth, non-thinking | 50.8% | ||||
| Zero-shot, Piano stem, medium | 50.4% | ||||
| Zero-shot, MIDI-text, non-thinking | 50% | ||||
| Zero-shot, Piano stem, non-thinking | 49.2% | ||||
| Zero-shot, No music, medium | 48.8% | ||||
| Zero-shot, Piano synth, thinking | 48.4% | ||||
| Zero-shot, Piano stem | 48.4% | ||||
| Zero-shot, MIDI-text, thinking | 47.6% | ||||
| Zero-shot, Piano synth, medium | 47.6% | ||||
| Zero-shot, MIDI-text, non-thinking | 47.2% | ||||
| Zero-shot, Piano synth | 47.2% | ||||
| GPT-5.6 Terra | Zero-shot, No music, high | 47.2% | |||
| Zero-shot, No music, medium | 47.2% | ||||
| Zero-shot, Piano stem, thinking | 46.8% | ||||
| Zero-shot, Piano stem, medium | 46.8% | ||||
| Zero-shot, No music, non-thinking | 46.4% | ||||
| Zero-shot, No music | 46.4% | ||||
| Zero-shot, Piano synth, thinking | 46.4% | ||||
| Zero-shot, MIDI-text | 45.6% | ||||
| Zero-shot, Piano stem, thinking | 45.2% | ||||
| Zero-shot, No music, non-thinking | 44.8% | ||||
| Zero-shot, No music, thinking | 44.8% | ||||
| Zero-shot, No music, thinking | 43.6% | ||||
| Zero-shot, No music, thinking | 42.4% | ||||
| Zero-shot, Piano synth | 38.8% | ||||
| Zero-shot, Piano stem | 37.2% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The PianoBench paper
Gemini 3.1 Pro on the official leaderboard passed 85% in February 2026
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official PianoBench 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
- 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 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (18 entries).
- Pass Models scored 16 models we track, plus 2 other entries on its leaderboard.
- Pass Newest score The newest is from September 2026, for GPT-6 Astra.
- Note Frontier models 1 of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-6 Astra (#4).
- Note Headroom Saturated. The best is 93.2%, past 85% since February 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 64 rows 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.