OpenAI Whisper large-v3 served on Groq's LPU hardware — top Whisper accuracy at Groq speed and cost, with word-level timestamps and automatic language detection.
12 languages · benchmarked
Overall Score
79.9
#11 of 35
Word Error Rate
14.41%
Character Error Rate
11.20%
Match Error Rate
13.34%
Word Info Lost
18.21%
Avg Latency
3.1s
Benchmarks Run
33
last 30 runs
Word error rate
14.41%
↓ 26.9% · 30d
3 runs ago
latest
Avg latency
3.1s
↑ 17.1% · 30d
3 runs ago
latest
overall WER · 35 models
Whisper Large V3
field · 34 models
lower-left is better
8 categories
WER
lower is better
Legal
2.7%
Medical
2.8%
Technical
5.3%
Finance
6.4%
General
8.1%
Conversational
10.9%
Noisy Environment
36.3%
Code-Switching
68.5%
5 accents
Rate
$0.0019
/min
Per-second billing. Bring your own provider key and pay your provider directly — a 5% routing fee applies (first 100 min/mo free).
Set up BYOKCost estimator
1 hour
$0.11
10 hours
$1.12
100 hours
$11.16
1,000 hours
$111.60
Billed per second of audio processed.
Model ID
groq/whisper-large-v3
Authenticate every request with your secret API key as a Bearer token. Issue a key from your dashboard.
Use it with your agent
Paste this into Claude Code, Cursor, Codex or any agent with a shell. It installs the CLI and the OpenTranscription skill, then transcribes with this model.
Quickstart
Upload your audio, create a job with this model, then poll the job or set a webhook_url to be notified when it completes.
Key parameters
file_path
string
required
Storage path returned by the upload step.
model
string
required
The model to run this job on.
models
string[]
Ordered fallback chain: primary first, then backups tried in order if a provider fails.
language
string
ISO 639-1 language code. Omit to auto-detect.
webhook_url
string
HTTPS URL notified with the result when the job completes. Delivered events are signed (X-OT-Signature).
title
string
Display name for this job in the dashboard. Falls back to the file name.
Response
A completed job returns the transcript in the OpenTranscription Unified Schema (OTUS).
Webhooks — skip polling
We POST a signed event to your webhook_url when a job completes or fails; verify the X-OT-Signature (HMAC-SHA256, reject if older than 300 s) and dedupe on the event id, then fetch the full transcript via GET /api/v1/transcriptions/{id}.
Rate-limited per tier — see the X-RateLimit-* response headers.
Full API referenceSupported Languages
Supported Formats
Features
Limits
Max file size
100 MB
Max duration
No documented limit
Whisper Large V3 is a speech-to-text model from Groq. In our standardized benchmarks it reaches 14.41% word error rate, ranking #11 of 35 models tested. It supports 12 languages and runs at $0.002/min — cheaper than most alternatives.
Its nearest benchmarked alternative is Soniox STT (Async). Best suited for high-volume, cost-sensitive pipelines.