Hi @buoyant halo. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
#What is gpt 4
abstract elbowBOT
GPT-4 is an older high-intelligence model developed by OpenAI. It is a large multimodal model, meaning it can process both text and (in some versions) image inputs to produce text outputs. GPT-4 was designed to solve complex problems with greater accuracy than previous models, leveraging advanced reasoning skills and a broad general knowledge base.
Key details about GPT-4:
- Context Window: 8,192 tokens (for the standard version)
- Modalities: Text input and output (image input is not supported in the standard API version)
- Knowledge Cutoff: December 1, 2023 (for the latest available version)
- API Availability: GPT-4 is available via the OpenAI API for paying customers, optimized for chat but also suitable for traditional completions tasks.
- Pricing: $30 per 1M input tokens and $60 per 1M output tokens (API usage)
- Features: Supports streaming and fine-tuning, but not function calling or structured outputs in the standard version
- Retirement in ChatGPT: As of April 30, 2025, GPT-4 was retired from ChatGPT and replaced by GPT-4o, but it remains available in the API GPT-4 model documentation, Sunsetting GPT-4 in ChatGPT.
GPT-4 marked a significant advancement in OpenAI’s model lineup, especially in advanced reasoning, complex instructions, and creativity, but has since been succeeded by newer models like GPT-4o for most use cases.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
buoyant halo
Nuh uh