#8282nd of 120 in Image and videoFor image, video and design, video and animation, science, health and research, education

code2video

Generates Manim educational videos from a topic using planner, coder and critic agents

Releases and problems feed (RSS)

Stars
2.4k
License
MIT
Last commit
Nov 2026
Language
Python
Stars, last 29 weeks+783
30 Apr 2026: 1,638 stars7 May 2026: 1,710 stars14 May 2026: 1,729 stars21 May 2026: 1,742 stars28 May 2026: 1,755 stars4 Jun 2026: 1,769 stars11 Jun 2026: 1,777 stars18 Jun 2026: 1,782 stars25 Jun 2026: 1,789 stars2 Jul 2026: 1,826 stars9 Jul 2026: 1,843 stars16 Jul 2026: 1,850 stars23 Jul 2026: 1,855 stars30 Jul 2026: 1,870 stars6 Aug 2026: 1,873 stars13 Aug 2026: 1,879 stars20 Aug 2026: 1,991 stars27 Aug 2026: 1,998 stars3 Sept 2026: 2,008 stars10 Sept 2026: 2,017 stars17 Sept 2026: 2,026 stars24 Sept 2026: 2,050 stars1 Oct 2026: 2,086 stars8 Oct 2026: 2,103 stars15 Oct 2026: 2,131 stars22 Oct 2026: 2,161 stars29 Oct 2026: 2,185 stars5 Nov 2026: 2,302 stars12 Nov 2026: 2,389 stars19 Nov 2026: 2,421 stars

1,638 on 30 Apr 20262,421 on 19 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Code2Video takes a knowledge point such as "Linear transformations and matrices" and produces an explainer video by having LLM agents write executable Manim code. A Planner expands the storyboard, a Coder synthesizes and debugs the code, and a Critic uses a vision-language model to refine layout. It ships with the MMMC benchmark of 117 topics and evaluation scripts.

Who it is for: Researchers and educators generating Manim explainer videos with LLM APIs

Strengths

  • Renders from Manim code, so output is editable and reproducible rather than pixel-generated
  • Separate Planner, Coder and Critic agents, with a VLM layout critique step
  • Includes MMMC benchmark (117 topics) with eval_TQ.py and eval_AES.py scripts
  • Single-topic and batch modes via run_agent_single.sh and run_agent.sh

Weaknesses

  • Requires paid external LLM and Gemini API keys; no local model path is described
  • Needs a separate Manim Community v0.19.0 install
  • Icon enrichment depends on IconFinder, which had API issues; a temporary icon set is provided
  • Shell scripts and JSON config only; no web UI, Docker image or tagged release

What it needs

  • Needs Manim Community v0.19.0, LLM API (Claude-4-Opus recommended), Gemini API key (VLM critic), IconFinder API key (optional)
  • Models: Claude-4-Opus, gemini-2.5-pro-preview-05-06

Also in Image and video

See all 120
Also in Image and video
RankProjectScore
1frigateLocal NVR with realtime object detection for IP cameras37.8k stars+950 (+2.6%) in 30 daysMIT94 out of 100
2toonflow-appInfinite-canvas tool for AI short drama and comic video production22.4k stars+3,446 (+18%) in 30 daysMIT89 out of 100
3MoneyPrinterTurboGenerates short videos from a topic, with script, footage, voiceover and subtitles144.9k starsMIT88 out of 100
#4InvokeAIWeb UI for diffusion image generation with inpainting canvas and node workflows29.1k starsApache-2.088 out of 100
#5autoclipTurns a video link into short clips with covers and captions12.5k stars+2,045 (+20%) in 30 daysMIT88 out of 100