OpenAI's Whisper Large V3 Turbo is an open-source audio model that transcribes speech to text and translates speech between languages. It is designed for automatic speech recognition and speech translation tasks.
Input
Output
Context
-
Max Output
-
Parameters
808.9M
Input Modalities
Output Modalities
None the catalog can show from stated facts or measured scores.
Download 1.62 GB · Runs on vLLM v0.32.0 · 17 GGUF builds
| Precision | Weights | Smallest GPU that fits |
|---|---|---|
| FP16 / BF16 | 1.51 GB | 1x NVIDIA T4 (16 GB) |
| INT8 | 0.75 GB | 1x NVIDIA T4 (16 GB) |
| INT4 | 0.38 GB | 1x NVIDIA T4 (16 GB) |
Computed by Inferbase from the 808,878,080 parameters in the creator's Hugging Face repository; weights only, as serving adds the framework and the input being processed.
Specification sources
As of Nov 14, 2026.
Answered from the facts on this page, with their sources.
OpenAI has not published a context window for Whisper Large V3 Turbo. The catalog does not estimate one.
Yes. Whisper Large V3 Turbo is an open-weight model under the MIT license. You can download and self-host it within the license terms.
Whisper Large V3 Turbo has 808,878,080 parameters (808.9M), as stated in the creator's Hugging Face repository.
Strengths: Open weights under a permissive license (mit); Supports 99 languages. Each is computed from the model's stated facts and benchmark results.
Whisper Large V3 Turbo accepts Audio and produces Text. The Capabilities card lists the API features it supports, such as function calling and structured output.
No benchmark source has published a result for Whisper Large V3 Turbo yet, so it is shown as unmeasured, not estimated.
Yes. Whisper Large V3 Turbo is available on Inferbase Managed through the OpenAI-compatible API as openai-whisper-large-v3-turbo-2. Request it by name, or request auto and routing picks it when it fits the request best.
The weights take 1.5 GB at FP16 / BF16, 0.8 GB at INT8, 0.4 GB at INT4, computed from the 808,878,080 parameters its creator publishes. Serving adds the framework and the input being processed.
The model card of Whisper Large V3 Turbo lists 99 languages: English, Chinese, German, Spanish, Russian, Korean, French, Japanese, Portuguese, Turkish, Polish, Catalan, Dutch, Arabic, Swedish, Italian, Indonesian, Hindi, Finnish, Vietnamese, Hebrew, Ukrainian, Greek, Malay, Czech, Romanian, Danish, Hungarian, Tamil, Norwegian, Thai, Urdu, Croatian, Bulgarian, Lithuanian, Latin, Māori, Malayalam, Welsh, Slovak, Telugu, Persian, Latvian, Bangla, Serbian, Azerbaijani, Slovenian, Kannada, Estonian, Macedonian, Breton, Basque, Icelandic, Armenian, Nepali, Mongolian, Bosnian, Kazakh, Albanian, Swahili, Galician, Marathi, Punjabi, Sinhala, Khmer, Shona, Yoruba, Somali, Afrikaans, Occitan, Georgian, Belarusian, Tajik, Sindhi, Gujarati, Amharic, Yiddish, Lao, Uzbek, Faroese, Haitian Creole, Pashto, Turkmen, Norwegian Nynorsk, Maltese, Sanskrit, Luxembourgish, Burmese, Tibetan, Filipino, Malagasy, Assamese, Tatar, Hawaiian, Lingala, Hausa, Bashkir, Javanese and Sundanese. That is what OpenAI states; the catalog does not test language quality.
Start on the free plan with 5,000 requests a month, no card required.