Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · Audio Flamingo 37 results · 7 benchmarks
Model card

Audio Flamingo 3.

unknown1 current SOTA
§ 02 · Benchmarks

Every benchmark Audio Flamingo 3 has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01AudioCapsAudio · Audio Captioningcider0.7%#1/1source ↗
02LibriSpeechSpeech · Speech Recognitionwer1.6%#3/92source ↗
03MMAUMultimodal · Audio-Text-to-Textaccuracy75.8%#4/4source ↗
04GigaSpeechSpeech · Speech Recognitionwer10.3%#37/47source ↗
05TED-LIUMSpeech · Speech Recognitionwer3.5%#42/50source ↗
06VoxPopuliSpeech · Speech Recognitionwer5.5%#55/55source ↗
07SPGISpeechSpeech · Speech Recognitionwer1.9%#56/56source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Audio Flamingo 3 actually performs.

Audio
1
benchmark
avg rank #1.0 · 1 SOTA
Multimodal
1
benchmark
avg rank #4.0
Speech
5
benchmarks
avg rank #38.6
§ 04 · Papers

1 paper with results for Audio Flamingo 3.

  1. 2025-07-10· 7 results

    Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
7
results
0 of 7 rows marked verified.