# danielgross/LlamaAcademy

A school for camelids

Repository: https://github.com/danielgross/LlamaAcademy
Canonical: https://ross.abutalabs.com/products/llamaacademy
Language: Python
License: MIT
License Family: permissive
Last push: 2023-05-01T10:35:41+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 88
- inputs: {"age_days": 1233, "days_push": 1220, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1198, forks 73 (observed 2026-08-28T04:03:57.738114+00:00)

## What it is
LlamaAcademy is a Python pipeline that crawls API documentation, generates synthetic training data with GPT-3.5/GPT-4, and fine-tunes a Vicuna-13B model with LoRA so it can write code against that API. The resulting mini-GPT can be self-hosted and used via LangChain for API glue-code generation.

## Use cases
- teach an LLM to call a specific API like Stripe or Notion
- fine-tune a local LLaMA model on API documentation
- generate synthetic instruction data from API docs
- host a custom mini-GPT that writes API integration code
- train a LoRA adapter on crawled documentation

## When to choose
- you want a bespoke model specialized in one API's documentation
- you have a >30GB VRAM GPU and an OpenAI API key
- you want to experiment with LoRA fine-tuning pipelines

## When to avoid
- you need consistently high-quality code generation - the project admits results are unreliable
- you lack GPU resources for fine-tuning
- you want a production-ready, maintained tool

## Facets
- artifact type: library
- maturity: experimental
- function: llm-training, rag, web-scraping, data-generation, agent-framework
- domain: large-language-models, machine-learning, artificial-intelligence, developer-tools
- platform: python
- tags: lora, fine-tuning, vicuna, llama, langchain, api-documentation, synthetic-data, openai, linux, gpu

## Member repositories
- danielgross/LlamaAcademy (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:57.738114+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 2026-08-30T06:21:09.284574+00:00, confidence not recorded.
  - readme: https://github.com/danielgross/LlamaAcademy (fetched 2026-08-28T04:03:57.738114+00:00, sha d00658344222)
- Data as of 2026-08-30T08:39:29.467469+00:00.