Hello macky.

I have a problem. I have entered my openai API, and remain report the following: Error: We couldn't generate the generated code. Please reload the page, or try again later.

Could you help me, thanks a lot.

Note that GPT4 costs 3 (input) to 6 (generated) cents for 1000 tokens, so you actually get more like 1 million tokens for $30.

Reference: https://openai.com/pricing

Your app is good and to some extent, brilliant.
However, it seems Streamlit starts rendering having only received partial code, and not the complete code for an animation.
This obviously, cause the animation not to render.
Here try this for example:
swing a double pendulum

In some cases, the scenes returned by the API can be mixed when there are multiple requests.

Example:

Code:

class GraphArrow(Scene):
    def construct(self):
        # Create axes and labels
        axes = Axes(x_range=[0, 10], y_range=[0, 10])
        x_label = MathTex(r'\text{Espaço}').scale(0.575)
        y_label = MathTex(r'\text{Tempo}').scale(0.75)
        labels = VGroup(x_label.next_to(axes.x_axis, RIGHT), y_label.next_to(axes.y_axis, UP))

        # Create dots
        d = Dot(color=WHITE, radius=0.1)
        d.move_to([0, 0, 0])
        d2 = Dot(color=WHITE, radius=0.1)
        d2.move_to([-1, 0, 0])

        # Group dots
        g = VGroup(d, d2)
        g.arrange()

        # Create traced paths
        trace = VMobject()
        trace2 = VMobject()

        trace_points = []
        trace2_points = []

        def update_trace(trace):
            trace_points.append(d.get_center())
            if len(trace_points) > 1:
                new_path = DashedVMobject(Line(trace_points[-2], trace_points[-1]), num_dashes=1, equal_lengths=True)
                trace.add(new_path)

        def update_trace2(trace2):
            trace2_points.append(d2.get_center())
            if len(trace2_points) > 1:
                new_path = DashedVMobject(Line(trace2_points[-2], trace2_points[-1]), num_dashes=1, equal_lengths=True)
                trace2.add(new_path)

        trace.add_updater(update_trace)
        trace2.add_updater(update_trace2)

        # Add elements to the scene
        self.play(Write(axes), Write(labels))
        self.add(d, d2, trace, trace2)
        self.wait(2)
        self.play(d.animate.move_to([0, 3, 0]), d2.animate.move_to([0, 2, 0]))
        self.wait(2)
        self.play(d.animate.move_to([0, -3, 0]), d2.animate.move_to([0, -2, 0]))
        self.wait(2)
        trace.clear_updaters()
        trace2.clear_updaters()

Video:

Hello macky.

These days I am thinking about how to make a more general path of text -> mainm code -> video.
Since the gpt code and weights are not open, I plan to train a model by myself.
We can work together if you want.

Engines

An engine is a complete text-to-code and code-to-video generation system for Generative Manim (GM). Since there are many ways we can think about video generation, we developed multiple engines so that they can be evaluated with LangSmith.

• GM powered by DSL
• GM powered by LangGraph
• GM OpenAI Fine-tuned
• GM OpenAI GPT-4 with Few-Shot Learning
• GM OpenAI GPT-4 with Zero-Shot Learning
• GM OpenAI GPT-4 with Ontologies/RAG

Anyone can suggest a new way (workflow, method, technique) to generate videos.

Platforms

API

Generative Manim will be available with a REST API.

• Create /code-to-video endpoint (Ready)
• Create /text-to-code endpoint (To be enhanced)
  • Add an option to switch between different GM engines (e.g. DSL, Ontologies, etc.)
  • Add first model (the simplest one, using GPT-4 with few-shot training and zero shot learning)

LangChain Integration

• Create a package over LangChain?

GPT / Chatbot

• OpenAPI YAML file with API endpoints

Streamlit

• New OpenAI API Key
• Integrate new system to detect possible errors on Streamlit

Testing

• Define how to test the results of GM engines, in order to choose the best. From the best engine that makes the best result, to the best engine with the lowest usage of tokens.