Side by side

See the difference.

Scores, capabilities and costs for the models you’re considering.

Scroll to compare models →

Comparison of OpenAI: GPT Audio Mini, abr-ai/niagara-19m-batch.en
2 modelsopenaiOpenAI: GPT Audio MiniRemove ×abr-aiabr-ai/niagara-19m-batch.enRemove ×
What’s differentA cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Input is priced at $0.60 per million...0.02B parameters · 1 language · Encoder: State-space model · Decoder: CTC
DateJan 19, 2026ListedOct 17, 2026Discovered
Listed rateSee provider quotesSee provider quotes
Provider endpoints0 endpoints →0 endpoints →
AMI-Cleaned WER ↗English short-form; normalization unversionedNot reported16.17% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
Earnings22-Cleaned-AA-chunked WER ↗English short-form; normalization unversionedNot reported16.1% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
Gigaspeech-Cleaned WER ↗English short-form; normalization unversionedNot reported12.72% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
LS Clean WER ↗English short-form; normalization unversionedNot reported4.74% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
LS Other WER ↗English short-form; normalization unversionedNot reported11.02% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
SPGISpeech WER ↗English short-form; normalization unversionedNot reported3.67% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
URGENT2024 WER ↗English short-form; normalization unversionedNot reported25.4% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
Voice Arena Monsoon WER ↗English short-form; normalization unversionedNot reported13.15% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
Voxpopuli-AA-Cleaned WER ↗English short-form; normalization unversionedNot reported5.73% WER ↗abr-ai/niagara-19m-batch.enLeaderboard entry
Context window128K—
Max output16.4K—
Parameters—0.02B
Tool callingSupportedNot reported
Structured outputSupportedNot reported
Audio inputSupportedSupported
TranscriptionNot reportedSupported

Add another model

Search all models →