Whisper Large v3 vs OpenAI Whisper (whisper-1)
Whisper Large v3 is an open-weight speech-to-text model you can call on this endpoint; OpenAI Whisper (whisper-1) is a closed model from OpenAI. Below: what each costs for three realistic workloads, the limits that apply, and when to choose which.
OpenAI figures are public list prices and documented limits, checked 2026-09-16 against the provider page. Inference APIs prices are read live from the price list.
- Cost: Whisper Large v3 is cheaper in all three workloads, by 67%.
- Limits: OpenAI: 25 MB per file. Rate limits scale with usage tier. Here: no per-minute or per-day caps.
Side by side
| Whisper Large v3 | OpenAI Whisper (whisper-1) | |
|---|---|---|
| Served by | Inference APIs | OpenAI |
| Model author | OpenAI | OpenAI |
| Model id | openai/whisper-large-v3 | whisper-1 |
| Price | $0.002 / audio minute | $0.006 / audio minute |
| Capabilities | Speech to text, 100 languages, Word timestamps, Speaker diarization, SRT / VTT | Speech to text, 99 languages, Timestamps, SRT / VTT |
| Weights | Open | Closed |
| Status | Available | Available |
| Limits | No per-minute or per-day request or token caps; usage draws on a prepaid balance | 25 MB per file. Rate limits scale with usage tier. |
| Speed | 0.047× real time (19.5 s clip in 0.92 s) | not measured by us |
What three workloads cost
List prices applied to the same work. The cheaper side of each row is in bold.
| Workload | Whisper Large v3 | OpenAI Whisper (whisper-1) | Difference |
|---|---|---|---|
| 1 hour of audio | $0.12 | $0.36 | 67% |
| 100 hours of audio | $12.00 | $36.00 | 67% |
| 1,000 hours of audio | $120 | $360 | 67% |
Whisper Large v3 is cheaper in all three workloads, by 67%.
When to choose which
- Cost is the deciding factor; it is cheaper in every workload above
- You want open weights: the same model can be moved to another host or self-hosted later
- Broad language coverage with automatic language detection
- Files up to 100 MB and SRT or VTT subtitles straight from the API
- Speaker labels with diarize=true, at the same price
- Your recordings exceed 25 MB and you do not want to chunk them
- You are already on OpenAI and your files are under 25 MB
Measured speed
Median of three runs on 2026-09-16: a 19.5-second mp3 clip, timed from upload to response. Expect ±30% with time of day. We did not measure OpenAI and do not quote other people's numbers; read "not measured" as unknown, not slow. Method and raw numbers: speed measurements.
Switching from OpenAI
Both speak the OpenAI wire format. The edit is the base URL, the key and the model id; then re-run your own prompts, because it is a different model.
import os
from openai import OpenAI
# before: OpenAI
# client = OpenAI(base_url="https://api.openai.com/v1", api_key=os.environ["OPENAI_API_KEY"])
# MODEL = "whisper-1"
# after: Inference APIs
client = OpenAI(base_url="https://api.inferenceapis.com/v1", api_key=os.environ["INFERENCE_API_KEY"])
MODEL = "openai/whisper-large-v3"
resp = client.audio.transcriptions.create(model=MODEL, file=open("meeting.mp3", "rb"))More detail, including Node, LangChain and LiteLLM: switching OpenAI-compatible providers.
FAQ
Which is cheaper, Whisper Large v3 or OpenAI Whisper (whisper-1)?
Whisper Large v3 is cheaper in all three workloads, by 67%. The table above uses list prices: Whisper Large v3 $0.002 / audio minute; OpenAI Whisper (whisper-1) $0.006 / audio minute.
Can I switch from OpenAI Whisper (whisper-1) to Whisper Large v3 without rewriting code?
Yes, if you call it through an OpenAI-compatible client. Change the base URL to https://api.inferenceapis.com/v1, swap the API key, and set the model to openai/whisper-large-v3. Because it is a different model, re-run your prompts and evaluations before moving production traffic.
What limits apply to OpenAI Whisper (whisper-1)?
25 MB per file. Rate limits scale with usage tier. On Inference APIs there are no per-minute or per-day request or token caps; usage draws on a prepaid balance.
Spotted a price or limit that has changed? Tell us and we will re-check the provider page.