OpenAI’s Post

OpenAI's latest innovation, GPT-4o mini, is here to change the game. 🎉 This cost-efficient small model is designed to make intelligence more accessible than ever, with prices starting at just 15 cents per million input tokens and 60 cents per million output tokens 💸. That's an order of magnitude more affordable than previous frontier models and over 60% cheaper than GPT-3.5 Turbo! 📉 So, what can you do with GPT-4o mini? 🤔 👉🏻Chain or parallelize multiple model calls: Perfect for applications that require multiple API calls or large volumes of context. 📊 👉🏻Build fast, real-time text responses: Ideal for customer support chatbots or any application that requires rapid text-based interactions. 💬 👉🏻Handle non-English text with ease: Thanks to the improved tokenizer, handling non-English text is now more cost-effective than ever. 🌎 👉🏻Supports text and vision inputs and outputs: With future support for image, video, and audio inputs and outputs. 📸 GPT-4o mini excels in various benchmarks, including textual intelligence, multimodal reasoning, math, and coding proficiency. 📊 It surpasses other small models in reasoning tasks, scoring 82.0% on MMLU, and demonstrates strong performance in function calling and long-context tasks. 📈

cost of input per MToken: 15 cents cost of output per MToken: 60 cents A 99% price reduction from text-davinci-003 back in 2022. Now it's getting interesting for scale.

The GPT-4o mini is here. Not expected this far the gpts have already come. The model's score compared with 77.9% for Google's Gemini Flash and 73.8% for Anthropic's Claude Haiku, of course according to OpenAI. Why it’s a big deal: we use lot of these models for some serous RnD! It’s something of value to EVERY industry because smaller /edge? language models require less computational power to run, making them a more affordable option for companies with limited resources looking to deploy generative AI in their operations. Note: With the mini model currently supporting text and vision in the application programming interface, OpenAI said support for text, image, video and audio inputs and outputs would be made available in the future.

This is actually revolutionary for anyone building with AI. A powerful model with Text, Voice, Image, and Video capabilities that is actually very cheap to use 🙌 this will open new opportunities for small businesses and organisations to use AI. One thing that openAI is great at is distribution, they know how to make their products available to a wide range of users.

AI advancements continue to push boundaries, and the introduction of GPT-4o Mini is no exception. We see how OpenAI's GPT-4o Mini offers a cost-efficient solution with outstanding performance. This small yet powerful model is ideal for a variety of applications, from customer support chatbots to handling large volumes of data. With its impressive benchmarks in textual intelligence and multimodal reasoning, GPT-4o Mini is a game-changer in AI accessibility and efficiency. #AI #ChatGPT #GPT4o

OpenAI’s new GPT-4o mini is designed to be a cost-effective AI model. It is priced at $0.15 per million input tokens and $0.60 per million output tokens1. This makes it significantly more affordable than previous models, including GPT-3.5 Turbo, which is more than 60% more expensive

OpenAI announces GPT-4o mini, a game-changer for affordable AI. This new model combines impressive performance with significantly reduced costs, making AI more accessible and affordable than ever before. Key highlights include: 🔹 82% score on MMLU benchmark 🔹 87% score on MGSM mathematical reasoning benchmark 🔹 60% cheaper than GPT-3.5 Turbo 🔹 Supports text and vision capabilities, with future updates for video and audio This could revolutionize AI adoption for SMBs. Thoughts on how this might impact your industry? Read the full article: https://lnkd.in/g88dhrcD #AI #OpenAI #TechNews #SmallBusinessAI

This is why building in AI is so different from regular software engineering. It’s like having an engineering team where the smartest engineers’ intelligence multiplies by an order of magnitude every 6 months while the average engineer keeps asking for a pay cut.

OpenAI’s latest announcement about GPT-4 Turbo promises a lot: more power, less cost, and an increasingly efficient AI model. Turbo is designed to be a powerhouse, handling complex tasks with ease and enabling smaller companies to scale AI applications more feasibly. However, we need to ask ourselves: Are these faster responses as accurate and insightful as before? Are we getting quality over quantity? Let’s not get swept away by the hype. While a cheaper, faster GPT is exciting, we must be critical of the broader impacts. Are we prepared for the rapid integration of such technology? And importantly, are the savings really worth it if we sacrifice the depth and quality of AI insights? In the end, GPT-4 Turbo may indeed be the next big thing in AI. But let’s keep our feet on the ground. After all, in the world of AI, the devil is always in the details.

While the LLMs lead the pack for innovation, the SLM market is the place to watch for rapid deployments and utilizations within the reach of more enterprises. Definitely a space to watch attentively.