# jingyaogong/minimind-v

👀 Train a 65M-parameter VLM from scratch in just 2h!

Repository: https://github.com/jingyaogong/minimind-v
Canonical: https://ross.abutalabs.com/products/minimind-v
Homepage: https://jingyaogong.github.io/minimind-v
Language: Python
License: Apache-2.0
License Family: permissive
Topics: artificial-intelligence, chatgpt, vision-language-model
Last push: 2026-08-06T14:08:24+00:00

## Health v2 (maintenance only)
Score: 61/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 96, release rhythm 21, longevity 51
- inputs: {"age_days": 721, "days_push": 27, "days_rel": 316, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8488, forks 932 (observed 2026-08-28T04:10:23.109039+00:00)

## What it is
MiniMind-V is an open-source educational project that trains tiny (26M-200M parameter) vision-language models from scratch in pure PyTorch, with a minimal 65M model trainable in ~2 hours on a single RTX 3090. It includes the full VLM stack—SigLIP2 vision encoder, MLP projector, LLM, plus dataset cleaning, pretraining, and SFT code—serving as both a minimal implementation and a tutorial for understanding multimodal models.

## Use cases
- train a vision-language model from scratch on a personal GPU
- learn how VLMs work by reading minimal code
- understand pretraining and SFT for multimodal models
- experiment with tiny multimodal models cheaply
- build an image-understanding chatbot
- study MoE and dense VLM architectures
- fine-tune a small VLM on custom image-text data

## When to choose
- you want to learn VLM internals with a transparent, minimal PyTorch implementation
- you have limited compute (single consumer GPU) and want to train a multimodal model end-to-end
- you need a small, OpenAI-API-compatible VLM for experimentation or education

## When to avoid
- you need state-of-the-art vision-language performance for production workloads
- you want a battle-tested enterprise multimodal solution rather than an educational codebase
- you need large-scale, high-resolution, or multi-image production features

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, llm-training, image-processing
- domain: artificial-intelligence, machine-learning, computer-vision, education
- platform: python, cross-platform
- tags: vision-language-model, vlm, multimodal, from-scratch, siglip, pytorch, educational, small-language-model, moe, sft, pretraining, natural-language-processing, gpu

## Member repositories
- jingyaogong/minimind-v (main) score 61

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:23.109039+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:26:49.946366+00:00, confidence not recorded.
  - readme: https://github.com/jingyaogong/minimind-v (fetched 2026-08-28T04:10:23.109039+00:00, sha 954e9909fa3e)
  - homepage: https://jingyaogong.github.io/minimind-v (fetched 2026-08-29T08:26:54.026724+00:00, sha cca98c2e2a7b)
- Data as of 2026-08-30T08:39:29.467469+00:00.