1sec.ai

Tag

#api-updates

Every item tagged api-updates, newest first.

5 items

modelsMay 7

Advancing voice intelligence with new models in the API

OpenAI has introduced new real-time voice models in their API, allowing for more natural voice interactions. These models can reason, translate, and transcribe speech. This advancement enables developers to build more intelligent voice experiences. You can integrate these models into your applications using the OpenAI API.

Key takeaways
  • New real-time voice models added to OpenAI API.
  • Models can reason, translate, and transcribe speech.
  • Enables more natural and intelligent voice experiences.
OOpenAI#voice-ai#api-updates
modelsApr 24

GPT-4 API general availability and deprecation of older models in the Completions API

OpenAI announced GPT-4 API general availability and a deprecation plan for older Completions API models, retiring at the start of 2024. GPT-3.5 Turbo, DALL·E, and Whisper APIs are also generally available. This change impacts builders using older models, who must migrate to newer versions or face API disruptions. You should review your API dependencies and plan for model upgrades.

Key takeaways
  • GPT-4 API is now generally available.
  • Completions API older models to retire early 2024.
  • GPT-3.5 Turbo, DALL·E, Whisper APIs also generally available.
OOpenAI#api-updates#model-deprecation#general-availability
modelsJan 25

New embedding models and API updates

OpenAI released new embedding models, including a smaller, more efficient text-embedding-3-small and a new text-embedding-3-large model. The updated models offer improved performance and pricing. You can use these models for tasks like clustering, semantic search, and anomaly detection.

Key takeaways
  • New text-embedding-3-small model is smaller and more efficient.
  • text-embedding-3-large offers improved performance.
  • Updated models support clustering, semantic search, and anomaly detection.
OOpenAI#embeddings#api-updates#performance
modelsNov 6

New models and developer products announced at DevDay

OpenAI announced several updates at DevDay, including GPT-4 Turbo with 128K context and lower prices, the Assistants API, GPT-4 Turbo with Vision, and DALL·E 3 API. These updates aim to improve performance, reduce costs, and expand functionality for developers. You can now access GPT-4 Turbo with a larger context window and lower pricing. The new Assistants API allows for more customized and automated interactions.

Key takeaways
  • GPT-4 Turbo now has 128K context window.
  • Assistants API is now available.
  • DALL·E 3 API is now available.
OOpenAI#api-updates#model-releases#developer-tools
modelsAug 22

GPT-3.5 Turbo fine-tuning and API updates

OpenAI introduced fine-tuning for GPT-3.5 Turbo, allowing developers to customize the model with their own data. This feature enables more accurate and relevant responses for specific applications. Fine-tuning also reduces the need for extensive prompt engineering. You can now improve model performance on your specific tasks.

Key takeaways
  • GPT-3.5 Turbo can be fine-tuned with custom data.
  • Fine-tuning reduces need for extensive prompt engineering.
  • Customization enables more accurate responses.
OOpenAI#fine-tuning#api-updates