OpenAI

Whisper Large V3

openai/whisper-large-v3

operationalZDRaudio

01·01

Overview

Modell-Beschreibung und Capability-Matrix.

About this model

Audio-Transkriptionsmodell von OpenAI (multilingual).

Capabilities matrix

streaming—

tools—

json—

vision—

reasoning—

embedding—

02·02

Providers

1 EU-Provider mit Region, ZDR-Status und Preisen pro Million Tokens.

Provider	Provider Model ID	Region	ZDR	Input €/M	Output €/M	Priority
Scaleway	whisper-large-v3	fr-par-1	ZDR	€0.50	—	100

03·03

Performance

Throughput, TTFT, E2E-Latency und Tool-Error-Rate — pro Provider, p50 über 24h.

Best Throughput

Scaleway188 tok/s

Lowest Latency

Scaleway1.81s

Scaleway

fr-par-1

Throughput

188tok/s

p50 · 24h

TTFT

261ms

p50 · 24h

E2E Latency

1.81s

p50 · 24h

Tool Err Rate

0.32%

last 7d

Note · Real metrics ship in Phase 4 once we log per-request TTFT + TPS into requests table aggregations.

04·04

Pricing

Pro Provider — relative Preisvisualisierung gegen den teuersten Anbieter dieses Modells.

Scaleway

ZDR

Input · per 1M tokens€0.50

Output · per 1M tokens—

Regionfr-par-1

05·05

Uptime

30-Tage-Heatmap pro Provider, aggregierte Verfügbarkeit und Incident-Historie.

Aggregated uptime

100.00%

Last 30 days · all providers combined

≥ 99 %95–99 %< 95 %

Scaleway

fr-par-1ZDR

100.00%

last 30 days

Recent incidents · last 30 days

No incidents affected this model in the last 30 days.

06·06

API

Drop-in OpenAI-kompatibler Endpoint. Tausche nur die baseURL — der Rest bleibt.

use-whisper-large-v3.ts

curl https://cleverouter.eu/v1/chat/completions \
  -H "Authorization: Bearer $CLEVERROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/whisper-large-v3",
    "messages": [
      { "role": "user", "content": "Hallo aus der EU." }
    ]
  }'