3 ms·

The naming convention is so messed up. o1, o3-mini (no o2, no o3???)

igravious 2y ago
https://www.perplexity.ai/search/new?q=list%20of%20all%20OpenAI%27s%20models https://www.perplexity.ai/search/new?q=list%20of%20all%20Ope... :) OpenAI has developed a variety of models that cater to different applications, from natural language processing to image generation and audio processing. Here’s a comprehensive list of the current models available: ## Language Models - \*GPT-4o\*: The flagship model capable of processing text, images, and audio. - \*GPT-4o mini\*: A smaller, more cost-effective version of GPT-4o. - \*GPT-4\*: An advanced model that improves upon GPT-3.5. - \*GPT-3.5\*: A set of models that enhance the capabilities of GPT-3. - \*GPT-3.5 Turbo\*: A faster variant designed for efficiency in chat applications. ## Reasoning Models - \*o1\*: Focused on reasoning tasks with improved accuracy. - \*o1-mini\*: A lightweight version of the o1 model. - \*o3\*: The successor to o1, currently in testing phases. - \*o3-mini\*: A lighter version of the o3 model. ## Audio Models - \*GPT-4o audio\*: Supports real-time audio interactions and audio generation. - \*Whisper\*: For transcribing and translating speech to text. ## Image Models - \*DALL-E\*: Generates images from textual descriptions. ## Embedding Models - \*Embeddings\*: Converts text into numerical vectors for similarity tasks. - \*Ada\*: An embedding model with various sizes (e.g., ada-002). ## Additional Models - \*Text to Speech (Preview)\*: Synthesizes spoken audio from text. These models are designed for various tasks, including coding assistance, image generation, and conversational AI, making OpenAI's offerings versatile for developers and businesses alike[1][2][4][5]. Citations: [1] https://learn.microsoft.com/vi-vn/azure/ai-services/openai/concepts/models [2] https://platform.openai.com/docs/models [3] https://llm.datasette.io/en/stable/openai-models.html [4] https://en.wikipedia.org/wiki/OpenAI_API [5] https://industrywired.com/open-ai-models-list-top-models-to-consider/ [6] https://holypython.com/python-api-tutorial/listing-all-available-openai-models-openai-api/ [7] https://en.wikipedia.org/wiki/GPT-3 [8] https://stackoverflow.com/questions/78122648/openai-api-how-do-i-get-a-list-of-all-available-openai-models/78122662
ben_w 2y ago
There's an o1-mini, there's an o3 it just hasn't gone live yet: https://openai.com/12-days/#day-12 https://openai.com/12-days/#day-12 they can't call it o2 because: https://en.wikipedia.org/wiki/The_O2_Arena https://en.wikipedia.org/wiki/The_O2_Arena and the venue's sponsor: https://en.wikipedia.org/wiki/O2_(UK) https://en.wikipedia.org/wiki/O2_(UK)
sumedh 2y ago
o3 will come later. o2 was not selected because there is already another brand with that name in UK

https://www.perplexity.ai/search/new?q=list%20of%20all%20OpenAI%27s%20models https://www.perplexity.ai/search/new?q=list%20of%20all%20Ope... :) OpenAI has developed a variety of models that cater to different applications, from natural language processing to image generation and audio processing. Here’s a comprehensive list of the current models available: ## Language Models - \*GPT-4o\*: The flagship model capable of processing text, images, and audio. - \*GPT-4o mini\*: A smaller, more cost-effective version of GPT-4o. - \*GPT-4\*: An advanced model that improves upon GPT-3.5. - \*GPT-3.5\*: A set of models that enhance the capabilities of GPT-3. - \*GPT-3.5 Turbo\*: A faster variant designed for efficiency in chat applications. ## Reasoning Models - \*o1\*: Focused on reasoning tasks with improved accuracy. - \*o1-mini\*: A lightweight version of the o1 model. - \*o3\*: The successor to o1, currently in testing phases. - \*o3-mini\*: A lighter version of the o3 model. ## Audio Models - \*GPT-4o audio\*: Supports real-time audio interactions and audio generation. - \*Whisper\*: For transcribing and translating speech to text. ## Image Models - \*DALL-E\*: Generates images from textual descriptions. ## Embedding Models - \*Embeddings\*: Converts text into numerical vectors for similarity tasks. - \*Ada\*: An embedding model with various sizes (e.g., ada-002). ## Additional Models - \*Text to Speech (Preview)\*: Synthesizes spoken audio from text. These models are designed for various tasks, including coding assistance, image generation, and conversational AI, making OpenAI's offerings versatile for developers and businesses alike[1][2][4][5]. Citations: [1] https://learn.microsoft.com/vi-vn/azure/ai-services/openai/concepts/models [2] https://platform.openai.com/docs/models [3] https://llm.datasette.io/en/stable/openai-models.html [4] https://en.wikipedia.org/wiki/OpenAI_API [5] https://industrywired.com/open-ai-models-list-top-models-to-consider/ [6] https://holypython.com/python-api-tutorial/listing-all-available-openai-models-openai-api/ [7] https://en.wikipedia.org/wiki/GPT-3 [8] https://stackoverflow.com/questions/78122648/openai-api-how-do-i-get-a-list-of-all-available-openai-models/78122662

ben_w 2y ago

There's an o1-mini, there's an o3 it just hasn't gone live yet: https://openai.com/12-days/#day-12 https://openai.com/12-days/#day-12 they can't call it o2 because: https://en.wikipedia.org/wiki/The_O2_Arena https://en.wikipedia.org/wiki/The_O2_Arena and the venue's sponsor: https://en.wikipedia.org/wiki/O2_(UK) https://en.wikipedia.org/wiki/O2_(UK)

sumedh 2y ago

o3 will come later. o2 was not selected because there is already another brand with that name in UK