[whatmodel.sh](../../README.md) / GPT-4o Audio Preview (2024-10-01) # GPT-4o Audio Preview (2024-10-01) OpenAI · released 2024-10 · proprietary **score 45.9** ±2.4 · comparison only: no current run option verified · from 10 benchmark results across 1 source Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks GPT-4o Audio Preview (2024-10-01) among every model on that task or test, worst to best.
other tests (10) 
test                                result  vs the others in its run  difficulty           counts  why
AHELM Speech Robust Bench (LibriS…   82.2%  ●·█··▆▄·▂▂·▄ beats 0/13   avg 34%                 44%  typical
AHELM CoRe-Bench conversational a…   34.2%  ▂·●··▂▂·▄▂█▄ beats 1/13   medium · avg 52%        23%  at its level
AHELM LibriSpeech transcription b…   11.3%  ▃···▃·●▃▅▃▅█ beats 3/13   avg 11%                 11%  typical
AHELM MUStARD audio sarcasm detec…   59.3%  ▂·······▆█●▆ beats 8/13   very hard · avg 57%    5.7%  blunt
AHELM AIR-Bench Foundation music …   52.7%  ▃●▃▅▃·▃·▃▅▃▃ beats 3/13   medium · avg 57%       4.2%  blunt · at its level
AHELM VoxCeleb2 same-speaker dete…   57.0%  ▂·▂·▂█▂●▄··▄ beats 8/13   very hard · avg 54%    4.1%  sharp · too hard for it · noisy
AHELM MELD audio emotion classifi…   45.6%  ▃··▃▃▃·▃▅●█▅ beats 7/13   medium · avg 39%       3.2%  blunt · at its level
AHELM FLEURS 7-language transcrip…   77.1%  ▄▂···●▆▂█▂·▂ beats 3/13   avg 83%                2.1%  blunt · noisy
AHELM Multilingual LibriSpeech 7-…   16.2%  ▂▂▂··▂●·▂▂█▆ beats 4/13   avg 14%                0.4%  blunt · noisy
AHELM VocalSound human-sound reco…   83.3%  ██▄▄·█·█·●▄▄ beats 10/13  easy · avg 67%         0.0%  blunt · at its level · noisy
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
AHELM: A Holistic Evaluation of Audio-L…       10   97% ▇▇▇▇▇▇▇▇▇▇▇▇
prior: pull toward the middle                        3% ▇
weighted average                                   = 45.9
Licences: AHELM: A Holistic Evaluation of Audio-Language Models ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) ## How to run it No verified API or local download yet. [interactive version](https://whatmodel.sh/model/gpt-4o-audio-preview-2024-10-01) · [JSON](https://whatmodel.sh/api?format=json&all=1)