audio model
Whisper v3 Turbo on Groq
GroqThe fast tier — cheapest path to a good-enough transcript.
Groq’s Whisper v3 Turbo is the "Fast" tier in Coworker. It is not the most accurate model in our harness and its tail latency (1197ms EN / 1273ms EL) is the slowest of the streaming options we measured, but it is cheap and dependable for everyday notes.
CapabilitiesVery low costWide language coverageOpen-weights lineage
Variants
1 way to run it.
Same family, different quality/cost rung. Picking the wrong one is the most common way teams overpay.
Turbo
Fastgroq-whisper-v3-turboFastest transcription — ideal for quick notes and everyday use.
1.2s
latency
1×
rel. cost
70
quality
Fit
Two audiences, one model.
The same model is a different proposition depending on what you point it at.
62/100
Recruitment fit
Fine for internal notes; not what you want under an interview a hiring decision rests on.
Best for
- Internal recruiter notes
- Quick voice memos
Strengths
- Cheap enough to leave on
- Broad language coverage
Watch out for
- Slowest tail latency of the streaming options we measured
- No speaker attribution
The numbers
Measured, not asserted.
HumanLike evals
Run on our own harness.
- Text still arriving after speaker stops (EN)
- 1197 msOur eval
Compare
