OpenAI has released an updated GPT-4o model with improved creative writing capabilities and better file handling.
Gpt-4o RSS
New strings of codes suggest OpenAI could be readying the vision capabilities in Advanced Voice Mode for a wider beta rollout.
Google's new experimental AI model, Gemini-Exp-1114, has outperformed OpenAI's models in many categories on Chatbot Arena, including vision, math, and creative writing.
DesktopGPT is an innovative new application that brings AI to the desktop with a focus on quick interactions and powerful templates that help everyone get the most out of AI.
OpenAI's new Predicted Outputs represents a significant step towards improving the user experience in LLM applications by addressing latency concerns.
OpenAI has released a new GPT-4o model with Structured Outputs and reduced pricing. Microsoft is also offering the new model at the same price, highlighting benefits such as consistent data formats.
If you want to have access to OpenAI's latest GPT-4o and 4o-mini on your desktop without using the browser, check out the just-updated DesktopGPT app from Stardock.
OpenAI has launched fine-tuning for its GPT-4o and GPT-4o mini models, enabling developers to customize them for specific use cases. This should result in improved performance and more.
OpenAI's new chatgpt-4o-latest model has significantly boosted ChatGPT's performance, leading to OpenAI reclaiming the No. 1 spot in the LMSYS Chatbot Arena with a record score of 1314.
OpenAI has released the system card and preparedness framework scorecard for its newest multimodal model, GPT-4o. It has implemented safeguards at both the model and system levels.
OpenAI released a new model, GPT-4o-2024-08-06, with Structured Outputs support and a reduced API price. The new model is now available on Microsoft Azure, and it seems to perform better.
OpenAI has released a new version of its multimodal model, GPT-4o, with reduced pricing. The new model is 50% cheaper for input tokens and 33% cheaper for output tokens compared to the previous model
OpenAI quietly announced that its GPT-4o model now supports a much larger output of 64,000 tokens, but at a higher cost. Microsoft also released the GPT-4o mini API with vision support.
Leaked benchmarks suggest that Meta's upcoming Llama 3.1 405B model could outperform OpenAI's GPT-4o on several key AI benchmarks, marking a potential milestone for open-source AI.
In response to the industry's trend of releasing cost-effective AI models, OpenAI announced GPT-4o mini. This new model is cheaper and faster than its predecessors, replacing GPT-3.5 Turbo.
A new study has outlined an attack using flowchart images with visual language models like GPT-4o to have it output unsafe outputs. The researchers even managed to automate the flowchart creation.
OpenAI has launched the ChatGPT Mac desktop app for free users. The app announced earlier this year alongside the GPT-4o model was only available to premium users of the AI chatbot until now.
A study has found that multimodel AI models perform poorly at giving safe responses when users give multimodal inputs such as an image and text together. The new SIUO benchmark was made as a result.
