Benchmarks · other
Hearing Bench
Models see only a spectrogram image of a sound clip, never the audio, and must name the sound: everyday sounds, instrument notes, vocal sounds, spoken words and sentences. Strict grading; value is % of 100 clips correct.
10 models · 38 results · updated 2026-10-28 · hearing-bench.adamholter.chatgpt.site ↗
Leaderboard
Gemini 4 Argon high
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 1 more models under All models in settings.
More from this benchmark
9 published measuresTime per task38 results · lower is better
1 more models under All models in settings.
Output tokens per task38 results · lower is better
Score vs any published measure7 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | ||||
|---|---|---|---|---|
| Gemini 4 Argon high | 15.0 | 8 | 6 | 15.0 |
| Opus 5.5 xhigh | 15.0 | 8 | 6 | 14.0 |
| Gemini 3.8 Flash high | 11.0 | 7 | 5 | 6 |
| GPT-6.1 Sol xhigh | 11.0 | 6 | 5 | 4 |
| GPT-6 Astra xhigh | 9 | 6 | 5 | 2 |
| Sonnet 5.5 xhigh | 9 | 5 | 4 | 2 |
| GPT-6.1 Terra high | 6 | 5 | 5 | 2 |
| Grok 4.7 high | 6 | 4 | 5 | 2 |
| Haiku 5.5 medium | 3 | 6 | 0 | 0 |
| GPT-6 Luna xhigh | 3 | 4 | 2 | 0 |
All results
38 rows| # | Model | Effort | Hearing Bench | Subscription $ / task | API $ / task | Time | Output tokens | Harness |
|---|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argon | high | 44.0% | $0.15 | $0.20 | 3.0 min | 20k | Antigravity |
| 2 | Opus 5.5 | xhigh | 43.0% | $0.003 | $0.19 | 1.7 min | 9.1k | Claude Code |
| 3 | Gemini 4 Argon | medium | 43.0% | $0.016 | $0.023 | 33s | 2.1k | Antigravity |
| 4 | Opus 5.5 | high | 39.0% | $0.001 | $0.043 | 23s | 1.9k | Claude Code |
| 5 | Opus 5.5 | medium | 35.0% | $0.000 | $0.026 | 13s | 1.0k | Claude Code |
| 6 | Opus 5.5 | low | 31.0% | $0.000 | $0.022 | 11s | 815 | Claude Code |
| 7 | Gemini 4 Argon | low | 30.0% | $0.005 | $0.007 | 14s | 498 | Antigravity |
| 8 | Gemini 3.8 Flash | high | 29.0% | $0.002 | $0.062 | 2.4 min | 16k | Antigravity |
| 9 | Gemini 3.8 Flash | medium | 27.0% | $0.001 | $0.029 | 68s | 7.6k | Antigravity |
| 10 | GPT-6.1 Sol | xhigh | 26.0% | $0.001 | $0.015 | 54s | 1.2k | Codex |
| 11 | GPT-6.1 Sol | high | 23.0% | $0.000 | $0.008 | 29s | 515 | Codex |
| 12 | GPT-6 Astra | xhigh | 22.0% | $0.002 | $0.053 | 42s | 805 | Codex |
| 13 | GPT-6 Astra | high | 21.0% | $0.001 | $0.029 | 23s | 338 | Codex |
| 14 | Sonnet 5.5 | xhigh | 20.0% | $0.000 | $0.013 | 11s | 1.0k | Claude Code |
| 15 | GPT-6.1 Sol | low | 20.0% | $0.000 | $0.004 | 17s | 192 | Codex |
| 16 | GPT-6.1 Terra | high | 18.0% | $0.000 | $0.001 | 7.0s | 169 | Codex |
| 17 | GPT-6 Astra | medium | 17.0% | $0.001 | $0.022 | 16s | 196 | Codex |
| 18 | Grok 4.7 | high | 17.0% | $0.010 | $0.11 | 5.9 min | 17k | Cursor |
| 19 | GPT-6 Astra | low | 16.0% | $0.001 | $0.021 | 17s | 173 | Codex |
| 20 | GPT-6.1 Sol | medium | 16.0% | $0.000 | $0.005 | 18s | 282 | Codex |
| 21 | GPT-6.1 Terra | medium | 16.0% | $0.000 | $0.001 | 7.0s | 135 | Codex |
| 22 | Sonnet 5.5 | high | 15.0% | $0.000 | $0.010 | 9.0s | 725 | Claude Code |
| 23 | Gemini 3.8 Flash | low | 15.0% | $0.000 | $0.005 | 20s | 1.1k | Antigravity |
| 24 | GPT-6.1 Terra | xhigh | 15.0% | $0.000 | $0.001 | 9.0s | 273 | Codex |
| 25 | Grok 4.7 | medium | 15.0% | $0.007 | $0.072 | 3.8 min | 11k | Cursor |
| 26 | Grok 4.7 | xhigh | 13.0% | $0.012 | $0.13 | 7.0 min | 22k | Cursor |
| 27 | Sonnet 5.5 | low | 11.0% | $0.000 | $0.006 | 6.0s | 331 | Claude Code |
| 28 | Sonnet 5.5 | medium | 10.0% | $0.000 | $0.007 | 7.0s | 455 | Claude Code |
| 29 | Grok 4.7 | low | 10.0% | $0.002 | $0.024 | 79s | 3.2k | Cursor |
| 30 | Haiku 5.5 | medium | 9.0% | $0.000 | $0.001 | 7.0s | 892 | Claude Code |