Compare model

openai / Transcription

OpenAIGPT-4o Mini Transcribe

openai/gpt-4o-mini-transcribe

Cost-efficient speech-to-text

Best for: Transcribing audio files into text

ShareLinkedInX

Catalog status

Enabled

Research checked

2026-10-06

Benchmark results

1 reported results

Results are source-attributed, not TokenRex measurements. Different tests and thinking variants are not interchangeable.

1 of 1 results

Independent multi-domain benchmark (OpenTranscription)

Accuracy · Third-party report

Test conditions and source

Third-party (non-AA, non-OpenAI) aggregate WER across 8 categories (medical, technical, legal, finance, general, conversational, code-switching, noisy); not an official OpenAI figure. Published source text: Independent benchmark site reports GPT-4o Mini Transcribe reaches 13.58% overall WER across its standardized multi-domain test suite, ranking #9 of 34 models tested.

Tested model: openai/gpt-4o-mini-transcribe

OpenTranscription — GPT-4o Mini Transcribe Benchmarks · Checked 2026-10-06

13.58% WER (lower is better)

Model specifications

Context window

16K

Published provider limit

Maximum output

2K

Published provider limit

Input and output modalities

Provider-reported formats; API compatibility is detailed below.

Input

Audio

Output

Text
Exact model ID
openai/gpt-4o-mini-transcribe
Model type
Transcription
Context window
16,000 tokens
Maximum output
2,000 tokens

A context window is the total conversation budget, not a separate maximum input allowance; generated output and reasoning can use that budget. Published provider limits are not independently tested TokenRex request limits.

Versions and thinking levels

One enabled version is currently listed for this model.

GPT-4o Mini TranscribeSelected$0.00302 per minute

Features and tool support

TokenRex’s exposed interface, not every feature advertised by the provider. “Available” describes the implemented interface, not a successful test of every input or tool.

Input through TokenRex

Available

Audio file upload (MP3, WAV, M4A and more, up to 25 MiB) on POST /v1/audio/transcriptions.

Output through TokenRex

Available

Transcript JSON, or words streamed live with stream=true (tested, including Hindi).

Function tools

Not exposed

Transcription endpoint only — it returns text and does not run tools.

Sources and verification

Provider specifications describe the upstream model, not a guarantee of every feature through TokenRex. Prices come from the enabled catalog; benchmark results belong to the exact tested model.