Official provider results
GPT Transcribe model documentation
OpenAI’s model page documents endpoints, features, and per-minute pricing for gpt-transcribe; it publishes no word-error-rate table on the page.
Source snapshots refreshed
Five measurement lanesVOICE / S2S
Reasoning, task completion, conversation dynamics, experience, and latency stay separate.
Benchmark profile
- Scope
- Endpoints, context features, and per-minute pricing from the official model page
- Primary metric
- Documented capabilities (no scored benchmark)
- Owner
- OpenAI
- Available evidence
- Specifications only
What it measures
Voice experience
Is the exchange natural, robust, and responsive?
Latency
How long do responses, tool calls, and full tasks take?
Available results
- Endpoints
- File and realtime transcription
- Context features
- Prompt, keywords, language hints
- Pricing
- $0.0045 per minute
Documented
v1/audio/transcriptions and v1/realtime/transcription_sessions.
Documented
Supports multilingual and code-switched audio without a fixed language.
Documented
Billed on audio duration.
Interpretation limit
This entry records what OpenAI documents for gpt-transcribe rather than a benchmark result. BenchLM assigns no score.