Tag
Manim is a Python-driven animation engine designed for math explanation videos, enabling precise control over LaTeX equations, geometric transformations, and 3D space animations. It is widely used in YouTube educational videos and academic presentations.
The author shares a quantization recipe for Qwen3.6 27B that makes the model use significantly fewer thinking tokens while still producing correct answers, leading to faster inference on math benchmarks.
A paper presenting SU-01, a 30B-A3B reasoning model that achieves gold-medal-level performance on IMO and IPhO problems via reverse-perplexity curriculum, two-stage reinforcement learning, and test-time scaling.
Benchmarked 7 LLMs on 5 math problems; Qwen3.5 27B and 35B A3B generated the longest reasoning chains, exceeding 10k tokens per question.
Manim is an open-source engine for creating precise programmatic math animations, originally developed by 3Blue1Brown. This repository hosts the original ManimGL version, while a separate community edition is maintained for broader collaboration.