# going-doer/Paper2Code

Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning

Repository: https://github.com/going-doer/Paper2Code
Canonical: https://ross.abutalabs.com/products/paper2code
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-03-25T01:00:05+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 73, release rhythm 35, longevity 35
- inputs: {"age_days": 495, "days_push": 162, "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 4925, forks 683 (observed 2026-08-28T04:09:02.548347+00:00)

## What it is
PaperCoder is a multi-agent LLM system that automatically generates a code repository from a machine learning research paper via a planning, analysis, and code generation pipeline. It also ships the Paper2Code benchmark datasets and model-based evaluation tooling for assessing generated repositories.

## Use cases
- generate a reference implementation from an arxiv paper
- reproduce a machine learning paper as code automatically
- convert a paper pdf or latex source into a runnable repo
- benchmark llm agents on paper-to-code generation
- evaluate generated repositories with model-based judges

## When to choose
- you want a starting-point implementation of an ML paper without writing it from scratch
- you need a benchmark for evaluating paper-to-code agents
- you want to run the pipeline with OpenAI models or self-hosted open-source models via vLLM

## When to avoid
- you need production-ready, fully correct code rather than a faithful scaffold
- you cannot share paper content with an external LLM API and cannot host your own model
- your paper is outside machine learning

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, machine-learning, rag
- domain: artificial-intelligence, large-language-models, machine-learning, developer-tools
- platform: python, cli, cross-platform
- tags: paper2code, multi-agent, llm, code-repository-generation, benchmark, papercoder, iclr, code-generation, ai-agents

## Member repositories
- going-doer/Paper2Code (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:02.548347+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 2026-08-29T18:18:00.270969+00:00, confidence not recorded.
  - readme: https://github.com/going-doer/Paper2Code (fetched 2026-08-28T04:09:02.548347+00:00, sha ba6a5f014fd6)
- Data as of 2026-08-30T08:39:29.467469+00:00.