Introduction
How often do you actually think and reason before you speak? The current state-of-the-art model, GPT-4o, was already delivering impressive responses without taking too long to respond. But imagine if it started taking longer to think and develop logic. With its latest model, o1, OpenAI has dropped a bombshell by introducing LLM models that can actually think and reason before responding – a capability that, until now, was considered exclusive to only a few humans.
OpenAI’s o1 series is a new series of ai models designed to spend more time thinking before generating an answer. Outperforming all previous versions and even many humans on various platforms such as the US Mathematical Olympiad (AIME), GPQA assessment, and Codeforces, the o1 series marks OpenAI’s significant step into ai. The two models, OpenAI o1 and OpenAI o1-mini, excel in reasoning, science, coding, and math.
So, in this blog, I decided to run some o1 experiments and put OpenAI o1 to the test. I ran three experiments involving physics, chemistry, and biology, combined with the magic of math and coding, to bring you the perfect dish prepared by o1.
Read on to discover the results of my experiments with OpenAI o1.
Overview
- Understand the scope of OpenAi's o1 model in the world of science, reasoning and logic.
- Visualize the possibilities with OpenAI's o1 in Physics, Chemistry, and Biology.
The power of images in education
The biggest challenge people face when it comes to science is the lack of visualization. Let’s imagine that for most of us the word “gravity” simply reminds us of an apple falling on Newton’s head. Visualization not only enhances the learning experience but helps us retain that lesson in our memory for a longer period. The idea of creating simulations/visualizations of scientific concepts is not really a novelty. But the power of creating these visualizations without writing a single line of code, to design interactive systems that go beyond following a set of rules while integrating logic and reasoning, is definitely new.
With Open ai's o1 model, I did exactly that!
I presented a physics, chemistry, and biology problem to OpenAI’s o1 model. The solutions to these problems required logical reasoning, mathematical calculations, and extensive coding, and o1 blew me away with the results!
Before moving on to the o1 experiments, I recommend you read our article on how to access OpenAI o1.
Experiment 1: Playing with planets (Physics)
Let me start with a quick refresher on our solar system: it consists of 8 planets: Mercury, Venus, Earth, Mars, Jupiter, Uranus, Saturn and Neptune. The Sun is at the center of our solar system and the planets revolve around it. Sounds simple, right?
Now, all of these planets are located at different distances from the Sun and revolve around it in unique orbits, at different speeds. The speed of a planet around the Sun is usually calculated using the following formula:
v = √(GM/r)
where:
- v is the speed of the planet
- G is the universal gravitational constant
- M is the mass of the Sun
- r is the radius or distance of the planet from the Sun
Let's say you want to visualize changes in a planet's speed by changing the planet's radius or the mass of the sun. All you need to do is ask OpenAI's o1 to write the code to create this visualization.
Immediate
I want to create a scientifically accurate simulation of our solar system with all 8 planets orbiting the Sun at their unique speeds. The simulation must include the following features:
- Adjustable parameters:
- Include sliders (drag bars) below the simulation to adjust the following for each planet and the Sun:
- The adjustment of the Sun's mass should affect the orbital velocities of the planets.
- Adjusting a planet's mass and radius should change its representation in the simulation (size and possibly color), but its own mass does not significantly affect its orbit due to the dominant mass of the Sun.
- Visual improvements:
- All planets and the Sun should be clearly labeled in the simulation with white text so that they are visible against the space background.
- The orbits of the planets should be shown as paths around the Sun.
- When a parameter is adjusted, the corresponding planet (or Sun) should be highlighted in the simulation for a short period (e.g. with a red rectangle) to indicate which celestial body was modified.
- User Interface:
- The text in front of each slider should be black for easy reading.
- The controls should be arranged in clear rows on a table, following the order of the planets in the solar system.
- For each celestial body, the format should be:
- Name of the planet or Sun
- Mass slider
- Radio slider
Production
Solar System Simulation
Solar System Simulation
Sun
Mercury
Click here to find the full code.
Things to Keep in Mind
To run this code, you just need to follow 3 steps:
- Copy this code into your favourite code editor, like Notepad.
- Save the File as index.html.
- Open the file in your favourite web browser.
Alternatively, you can directly play with the version I created. Do share in the comments what happened to jupiter’s speed when you maxed out on its radius?
Working of OpenAI’s o1 in this Experiment
While you marvel at this application, lets take a step back to understand what did OpenAI’s latest model did behind the scenes to bring my visualisation to life.
- It took sometime to understand the different aspects that it needs to consider by groing through the prompt.
- It realised that it needs to use the concepts related to physics, mathematics & coding to generate the output.
- It combined the logic behind each step – merging physics & mathematics and translated several visual elements that i had suggested into a suitable code.
Now, that we are done with modeling our Solar system, let’s get some chemicals brewing.
Experiment 2: Acid-Base – Visual Chemistry 🧪
There are thousands of acids and bases out there. It’s not always easy to remember which one of these reacts with each other and the chemical they create? Imagine if we knew the results that we could get before mixing two chemicals! It would probably save us from many burns or unfortunate accidents in the lab and might as well help our institutes save money over broken beakers and other equipment.
So my ask to Open ai o1 was to create a simulation in which we could pick an acid, a base, and their quantities and it would tell us how our product would look like.
Prompt
Create a dynamic and interactive simulation involving three labeled beakers:
- Beaker Descriptions:
- Beaker A: Contains a selected acid.
- Beaker B: Contains a selected base.
- Beaker C: Shows the output of the reaction.
- User Interface Elements:
- Include dropdown menus for selecting 20 different acids and 20 different bases.
- Provide separate dropdown menus to choose volumes (from 10 to 100 mL, in increments of 10) for both acid and base.
- Interactive Functionality:
- When the user selects an acid, base, and their volumes, the simulation should:
- Animate the addition of the acid and base into their respective beakers.
- Display the acid and base labels on Beaker A and Beaker B once selections are made.
- Show a green color in Beaker C if a reaction occurs and a blue color if no reaction occurs.
- Provide detailed information below the simulation, indicating whether a reaction occurred, the product generated, its name, chemical formula, and relevant details.
- Visual Elements:
- Beakers should have a realistic shape.
- Ensure all text is in black for readability.
Output
Acid-Base Reaction Simulation
Click here to find the full code.
Things to keep in mind
To run this code, follow the same steps mentioned above.
Alternatively, you can use directly The version I created.
Now that our chemistry is sorted, it's time to move on to our next o1 experiment!
Experiment 3: Mixing with biology
The only thing that separates us from machines is biology. Biology holds the entire secret of humanity, and at the core of biology are proteins. Proteins are to humans what tokens are to LLMs. These proteins make up our body, our brain, and our entire nervous system, helping us understand and comprehend our environment. This is similar to how tokens build the entire functionality of LLMs.
But the possible combinations of proteins are virtually unlimited, so it can be very difficult to remember the names and uses of each one.
So the task I gave to OpenAI's o1 was to create a simulation that could help me generate unlimited combinations of these proteins and learn their use cases.
Immediate
Create an interactive protein creation simulation with the following features:
- User Interaction:
- Provides a drop-down menu containing the 20 standard amino acids, showing their full names, three-letter codes, and one-letter symbols.
- Includes buttons to add amino acid to the chain, delete the last amino acid, and reset the chain.
- Visual representation:
- Start with the most basic amino acid, glycine, which is shown by default.
- Represent each amino acid as a single-colored helix and show their one-letter symbols below.
- Visually connect amino acids with lines or bonds to represent peptide bonds as the chain grows horizontally.
- Information display:
- As amino acids are added, display their names and basic information (properties, uses) below the simulation.
- If the amino acid sequence matches a known protein or peptide, it displays detailed information, including its name, description, and popular uses.
- For sequences that do not match known proteins, show the amino acid sequence and general information about the peptide, indicating that it may represent a new or synthetic peptide.
Production
Protein Builder Simulation
Click here to find the full code.
To run this code, follow the same steps mentioned above.
Alternatively, you can use directly The version I created.
This is amazing! I never imagined that identifying proteins and creating new ones could be so fun and easy.
My observation from the previous experiments
This latest o1 model impresses with its reasoning, logical thinking, and coding capabilities. However, it needs to make significant progress to add features available in GPT4o, such as web browsing, file uploads, or working with images. Until we see those improvements in the o1 model, GPT 4o will remain the reference model for common tasks.
If you want to learn more about how OpenAi's o1 and o1-mini work, read these articles:
Conclusion
I'm blown away by the results I've seen from the o1 experiments above. With just a couple of hours, I was able to create 3 simulations for 3 different areas! o1 can potentially help millions of students who don't have the resources to truly experience the possibilities that science has to offer. It will be immensely beneficial to anyone who has an idea and wants the world to see it.
Although in the current version of the model we cannot add images or audio files, when that happens, this multimodality will further expand the possibilities that can be achieved with this model. A truly generative future awaits us.
Stay tuned to Analytics Vidhya's blog to know more about the uses of o1!
Frequently Asked Questions
Q1. What is o1?
A. OpenAI o1: A new series of ai models designed to spend more time thinking before responding. These models can reason about complex tasks and solve more difficult problems than previous models in science, programming, and mathematics.
Q2. When was OpenAI's o1 model launched?
A. OpenAI's o1 model was launched on September 12, 2024.
Q3. Can the o1 model process images?
A. Yes, the latest o1 models can process images, although this functionality is not yet available to the public.
Q4. Can everyone use the o1 models?
A. Currently only paid members can use OpenAI's o1 model.
Q5. How is OpenAI's o1 series trained?
A. The o1 series is trained with large-scale reinforcement learning that allows it to reason using chains of thought.
Anu Madan has over 5 years of experience in content creation and management. She has worked as a content creator, reviewer, and manager and has created several courses and blogs. Currently, she is working on creating and strategizing content curation and design around generative ai and other emerging technologies.
Frequently Asked Questions
Customer is very important, customer will be followed by customer,
Privacy and Cookies Policy
(tags to translate)o1
