# rasbt/llm-architecture-gallery

LLM Architecture Gallery source data

Repository: https://github.com/rasbt/llm-architecture-gallery
Canonical: https://ross.abutalabs.com/products/llm-architecture-gallery
Homepage: https://sebastianraschka.com/llm-architecture-gallery/
License: Apache-2.0
License Family: permissive
Topics: ai, llms
Last push: 2026-08-16T20:37:36+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 35, longevity 12
- inputs: {"age_days": 171, "days_push": 17, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1495, forks 130 (observed 2026-08-28T04:04:53.091550+00:00)

## What it is
Source metadata (models.yml) for Sebastian Raschka's LLM Architecture Gallery, a curated reference of ~99 modern language-model architectures with diagrams, fact sheets, and comparison tools. This repo is the lightweight public data export, not the website itself.

## Use cases
- compare llm architectures side by side
- look up attention mechanisms used by a model
- find release dates and parameter counts for llms
- browse a curated list of modern language model architectures
- study how llama or qwen architectures differ
- find implementation links for open llm models

## When to choose
- you need structured metadata about LLM architectures in a machine-readable format
- you want to build your own visualization or tool on top of the gallery data
- you're researching architecture trends across many LLM families

## When to avoid
- you want the full gallery website implementation or rendering code
- you need multimodal model component details rather than text backbones
- you need runnable model code or weights

## Facets
- artifact type: dataset
- maturity: active
- function: machine-learning, llm-inference, data-visualization, documentation
- domain: large-language-models, artificial-intelligence, education
- platform: -
- tags: llm-architectures, reference, metadata, yaml, curated, web-server

## Member repositories
- rasbt/llm-architecture-gallery (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:53.091550+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 2026-08-30T04:33:18.778406+00:00, confidence not recorded.
  - readme: https://github.com/rasbt/llm-architecture-gallery (fetched 2026-08-28T04:04:53.091550+00:00, sha 8cb703433a67)
  - homepage: https://sebastianraschka.com/llm-architecture-gallery/ (fetched 2026-08-29T11:39:09.677721+00:00, sha 1064853e05c8)
  - site_page: https://sebastianraschka.com/about (fetched 2026-08-29T11:39:09.687720+00:00, sha 667cdfd072dc)
  - site_page: https://sebastianraschka.com/llm-architecture-gallery/changelog (fetched 2026-08-29T11:39:09.690811+00:00, sha 905d6edb266d)
- Data as of 2026-08-30T08:39:29.467469+00:00.