Of all the AI chatbot comparisons here on PC Guide, ChatGPT vs Amazon Olympus AI may prove to be the most brutal bout of all. Without even a confirmed release date, the speculative name of an Amazon AI model is drawing buzz with a rumored “two trillion parameters”, which is double that of OpenAI’s ChatGPT foundation model, GPT-4. We speculate on how these two AI models will compare, updating our conclusions as confirmed information becomes available.
ChatGPT vs Amazon Olympus AI – rumored performance
Amazon Olympus is a rumored AI model, with no confirmed release date. As reported by Reuters, “the people” who were the source of this information “spoke on condition of anonymity because the details of the project were not yet public”. The name of the chatbot that would feature this speculative model has not been revealed. Amazon currently has no consumer-level AI chatbot, but does operate Amazon Lex, a platform that allows businesses to “build bots with Conversational AI”.
In terms of raw power, measured in model parameters, Amazon Olympus could dethrone OpenAI’s ChatGPT. That is an extremely bold claim, considering that OpenAI has held a substantial lead across the big tech landscape since the release of ChatGPT in November 2022. However, if it’s true that this model “has 2 trillion parameters,” as “the people said”, then we’ll know how it ranks against ChatGPT’s GPT-4 model, which reportedly has “1.76 trillion parameters” by comparison.
This is in no small part because of the $13 Billion USD in funding that Microsoft has invested into the ‘underdog’ since 2019, and so comparing the funding of OpenAI to Amazon is arguably like comparing Microsoft to Amazon – a much fairer fight. Amazon is more publicly working on other new artificial intelligence endeavors such as generative AI for AWS and Alexa. Along with Olympus, these tools could be used to generate high-quality background product images and product descriptions.
Ultimately, we can’t compare ChatGPT vs Amazon Olympus until the latter becomes available. Amazon’s AI advancements with this new AI model sound pretty cool, but it’s not even in closed Beta yet – it’s still being trained. We can safely recommend you focus on ChatGPT for now.
| Company | CEO | AI Chatbot | LLM | API | Open-source |
|---|---|---|---|---|---|
| xAI | Elon Musk | Grok | Grok-1 | No | No |
| OpenAI | Sam Altman | ChatGPT | GPT-3.5, GPT-4, GPT-4V, or GPT-4 Turbo | Yes | No |
| Sundar Pichai | Bard | PaLM 2 | Yes | No | |
| Microsoft | Satay Nadella | Bing Chat | GPT-4 | No | No |
| Meta | Mark Zuckerberg | Meta AI | LLaMA 2 | No | Yes |
| Anthropic | Dario Amodei | Claude | Claude-2 | Yes | No |
| Amazon | Andy Jassy | N/A | Olympus (rumored) | No | No |
Essential AI Tools
Content Guardian – AI Content Checker – One-click, Eight Checks
Jasper AI
WordAI
Copy.ai
Writesonic
AI model benchmarks compared
Aside from money and availability, it’s worth comparing tech to tech. Measuring the performance of these popular LLMs can be done in a few different ways.
Just like computer hardware, we have benchmarks for this type of thing. One such number we can use to compare two AI models is their parameter count. However, until benchmarking data is shared from Apple itself, all we know are the ranking of other models in the AI space.
| Benchmark | Grok-0 | LLaMa 2 | Inflection-1 | GPT-3.5 | Grok-1 | PaLM 2 | Claude-2 | GPT-4 |
|---|---|---|---|---|---|---|---|---|
| GSM8k | 56.8% | 56.8% | 62.9% | 57.1% | 62.9% | 80.7% | 88% | 92% |
| MMLU | 65.7% | 68.9% | 72.7% | 70.0% | 73.0% | 78% | 75% | 86.4% |
| HumanEval | 39.7% | 29.9% | 35.4% | 48.1% | 63.2% | N/A | 70% | 67% |
| MATH | 15.7% | 13.5% | 16.0% | 23.5% | 23.9% | 34.6% | N/A | 42.5% |
The actual technology behind these two Chatbots is a type of AI model called an LLM. That stands for Large Language Model, and it’s a neural network trained solely on text data (the dataset). Natural language processing (NLP) is a sub-field of artificial intelligence and is an LLM in action. NLP is the ability of a computer to communicate with a human, using realistic human language.
FAQ – Could Amazon Olympus be more powerful than ChatGPT?
While we can’t benchmark Amazon’s Olympus AI model yet, we can compare the reported parameter count of “2 trillion” to GPT-4‘s “1.76 trillion“. All we can say at this point is that higher parameter count doesn’t necessarily mean higher performance, and doesn’t take many important factors into account. Factors like censorship, internet access, multimodality, plugins, and API support all make a difference beyond the big exciting parameter number.
In all these factors, ChatGPT is already a mature piece of software with dozens of AI capabilities. This means that Apple will have to beat OpenAI in not just parameter count, but in multimodality and features as well. However, while we remain reserved about our conclusion on Amazon Olympus, there’s no doubt that Amazon will be turning heads with AI tech this year.
