Downloads · 30 days
0
sebastianboehler/autoresearch-manim-qwen25coder-3b-mlx-lora
autoresearch-manim-qwen25coder-3b-mlx-lora is a text generation model from sebastianboehler. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as other.
This repository contains an MLX LoRA adapter for Qwen/Qwen2.5-Coder-3B-Instruct, tuned for Manim Community Edition code generation.
Downloads · 30 days
0
Access
Public
Updated Apr 5, 2026
Repo size
39.9 MB
Likes
0
Public
Click a slice to open those files.
.safetensors20 MB · 100%
From the Hugging Face model README
This repository contains an MLX LoRA adapter for Qwen/Qwen2.5-Coder-3B-Instruct, tuned for Manim Community Edition code generation.
The adapter is trained to produce runnable educational animation scenes, with a bias toward:
This is an adapter-only release, not a merged full checkpoint.
Qwen/Qwen2.5-Coder-3B-Instruct128307242render-first600.529The adapter was trained on the curated public dataset:
17214428129 / 21 / 22The corpus focuses on high-quality Manim code samples that were either hand-authored, imported from documentation and repositories, or promoted through manual review after successful render checks.
Held-out evaluation for this promoted release completed on 2026-04-06 using the repo's repair-aware generation-plus-render harness on the 22-case gold-silver test split.
0.8180.7220.7090.5541.740Interpretation:
Important:
mlx_lm.generate alone.Clone or download this adapter repo locally, then run it with mlx-lm against the base model:
git clone https://huggingface.co/sebastianboehler/autoresearch-manim-qwen25coder-3b-mlx-lora
python -m mlx_lm.generate \
--model Qwen/Qwen2.5-Coder-3B-Instruct \
--adapter-path ./autoresearch-manim-qwen25coder-3b-mlx-lora \
--system-prompt "You write runnable Manim Community Edition Python files. Return only Python code, use from manim import *, and define exactly one scene class." \
--prompt "Create a 10-second Manim scene that explains gradient descent on a contour plot." \
--max-tokens 1600 \
--temp 0.0 \
--top-p 0.9 \
--seed 42
adapters.safetensors: release adapter weightsadapter_config.json: MLX LoRA training configurationbest_checkpoint.json: restored best-step metadatarelease_metadata.json: release-level dataset and training summary