What are the languages supported by GPT-4 and GPT-4o? Is there data on which are the langauges with the most training data, excluding English?
#GPT-4 and GPT-4o language support
thorny ivy
pseudo mistBOT
Hi @thorny ivy. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
GPT-4o and GPT-4 support a wide range of languages. The list includes but is not limited to:
- Albanian
- Amharic
- Arabic
- Armenian
- Bengali
- Bosnian
- Bulgarian
- Burmese
- Catalan
- Chinese
- Croatian
- Czech
- Danish
- Dutch
- Estonian
- Finnish
- French
- Georgian
- German
- Greek
- Gujarati
- Hindi
- Hungarian
- Icelandic
- Indonesian
- Italian
- Japanese
- Kannada
- Kazakh
- Korean
- Latvian
- Lithuanian
- Macedonian
- Malay
- Malayalam
- Marathi
- Mongolian
- Norwegian
- Persian
- Polish
- Portuguese
- Punjabi
- Romanian
- Russian
- Serbian
- Slovak
- Slovenian
- Somali
- Spanish
- Swahili
- Swedish
- Tagalog
- Tamil
- Telugu
- Thai
- Turkish
- Ukrainian
- Urdu
- Vietnamese
You can find the full list in the OpenAI Help Article (https://help.openai.com/en/articles/8357869-how-to-change-your-language-setting-in-chatgpt#h_513834920e).
As for the languages with the most training data, excluding English, the exact details are not provided in the knowledge sources. However, it is mentioned that the performance on transcription in a given language is directly correlated with the amount of training data employed in that language. This suggests that languages with more training data would have better performance. You can find more details in the Whisper Model Card (https://github.com/openai/whisper/blob/main/model-card.md).
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
thorny ivy
@pseudo mist the link provided is for ChatGPT, not the API models. Do the API GPTs support the same list of languages?
pseudo mistBOT
Let me check the knowledge base for you. One moment! :books:
Yes, the GPT-4 and GPT-4o models, which are available through the OpenAI API, support a wide range of languages. However, the exact list of supported languages for these models is not explicitly mentioned in the provided knowledge sources.
It's important to note that while these models have been trained on diverse internet text and can generate text in various languages, their performance can vary depending on the language due to the amount and quality of the training data in that particular language.