Quick Answer

Papers with Code is gone. Here's what to use instead.

Status:
Shut down by Meta in July 2025, redirects to Hugging Face
What was lost:
9,327 benchmarks, 79,817 papers, SOTA leaderboards
Best alternative:
CodeSOTA - independent, verified benchmarks, updated December 2025
Data archive:
GitHub: paperswithcode/paperswithcode-data (historical, not updated)
The Story

What Happened to Papers With Code?

For seven years, Papers with Code was the definitive resource for ML research. Then Meta shut it down without warning. Here's what was lost, what remains, and why CodeSOTA exists.

December 2025|8 min read

Timeline

July 2018Robert Stojnic and Ross Taylor launch Papers with Code
Dec 2019Meta (Facebook AI) acquires for ~$40M, promises to keep it open
Oct 2020arXiv integration adds "Code" tab to paper pages
Peak79,817 papers, 9,327 benchmarks, 5,628 datasets indexed
July 2025Meta shuts down the site without notice. Redirects to Hugging Face.

What Papers With Code Was

Papers with Code was "the Wikipedia of machine learning research." Founded by Robert Stojnic (one of Wikipedia's original developers) and Ross Taylor, it solved a fundamental problem: connecting research papers with working code implementations.

Before Papers with Code, finding an implementation of a paper meant:

  • Searching GitHub with random keyword combinations
  • Hoping the authors released code (most didn't)
  • Finding outdated implementations in deprecated frameworks
  • Spending days reproducing papers from scratch

Papers with Code changed this. Every paper had linked implementations (official and community), benchmark results, dataset information, and method explanations. The wiki model let anyone contribute, building a comprehensive knowledge base that became essential infrastructure.

The Benchmark System

The most valuable contribution was the benchmark tracking system. Every benchmark was organized as a <Task, Dataset, Metric> tuple:

# Example benchmark
Task: Image Classification
Dataset: ImageNet
Metric: Top-1 Accuracy

Content was organized into 16 research areas covering everything from Computer Vision to Robotics. Each area contained tasks, sub-tasks, and specific benchmarks with leaderboards showing:

  • Model rankings with metric scores
  • Links to papers and code implementations
  • Visual timelines showing SOTA progression over years
  • Official vs community implementation badges

Why It Mattered

For Researchers

Establish baselines instantly. Know what SOTA looks like before starting a project. Find existing implementations to build on rather than reimplementing from scratch.

For Engineers

Find working code for papers. Compare frameworks (PyTorch vs TensorFlow implementations). Skip the "reproducing papers" phase and go straight to production.

For the Field

Research shows papers with linked code get ~20% higher citation rates. Papers with Code made reproducibility the norm, not the exception.

For Decision Makers

Clear benchmark comparisons across models. Understand what's actually state-of-the-art vs marketing claims. Make informed technology choices.

"Papers with Code was bread and butter for Research Engineers and Scientists."- ML community sentiment

The Shutdown

On July 24-25, 2025, Meta "sunsetted" Papers with Code without prior notice. Users reported "Bad Gateway 502" errors and garbled text. GitHub issues went unanswered. The site now redirects to Hugging Face's "Trending Papers" feature.

What Was Lost

  • - Comprehensive SOTA leaderboards across 9,327 benchmarks
  • - Paper-to-code linkages for 79,817 papers
  • - Method explanations and connections
  • - The unified research workflow millions relied on

The irony: when Meta acquired Papers with Code in 2019, they promised it would "remain a neutral, open and free resource." That promise lasted five and a half years.

The Vacuum Left Behind

Hugging Face's successor provides paper discovery and code links, but lacks the comprehensive SOTA leaderboards that defined Papers with Code. Hugging Face focuses on model-centric leaderboards (like the Open LLM Leaderboard), not paper-centric benchmark tracking.

FeaturePapers With CodeHugging Face
Paper discoveryYesYes
Code linksYesYes
SOTA leaderboards9,327 benchmarksLimited
Dataset registry5,628 datasetsDifferent focus
Method explanationsYesNo
Task hierarchy16 areas, nestedNo

The current alternatives (Semantic Scholar, Connected Papers, Kaggle) each cover parts of what Papers with Code did, but none replicate the unified paper-code-benchmark-dataset-method linkages. The integrated experience is gone.

What Was Saved

The community moved quickly to preserve data:

  • GitHub archives - Historical JSON dumps at paperswithcode/paperswithcode-data
  • Hugging Face datasets - pwc-archive with papers, abstracts, evaluation tables
  • ORKG - Imported 2021 benchmark data into Open Research Knowledge Graph

The data exists. The integrated experience doesn't.

Why CodeSOTA Exists

CodeSOTA is building what Papers with Code provided: verified benchmarks, practical recommendations, and runnable code. We're starting focused (OCR and document AI) rather than trying to index everything at once.

Our approach is different in key ways:

  • Verified results - We don't just aggregate claims. We run the benchmarks ourselves where possible.
  • Practical focus - Not just leaderboards. Which model for your use case? What are the real tradeoffs?
  • Open data - All benchmark data available as JSON. Build on it, cite it, contribute to it.
  • Independent - Not owned by a big tech company that might shut it down.

Why CodeSOTA as a Papers with Code Alternative

CodeSOTA starts where Papers with Code left off. We're building on the PWC dataset foundation and adding what was missing: fresh benchmarks, verified results, and community-driven updates.

The flywheel:

1
Seed data

PWC's 79K papers and 9K benchmarks as the starting point

2
Fresh verification

We run benchmarks ourselves, flag stale claims

3
Community submissions

Researchers and vendors submit new results

4
Continuous improvement

Feedback drives better coverage and accuracy

What's live now:

  • - OCR benchmarks: 16+ models, 9 datasets, verified December 2025
  • - Model tutorials and comparison pages
  • - Open JSON API at /data/benchmarks.json
  • - Paper submission workflow for new results

Coming soon:

  • - Speech recognition, code generation, computer vision
  • - Automated benchmark verification pipeline
  • - Community voting on result accuracy
  • - LLM evaluation comparisons

The continuation, not just an alternative:

  • - Built on PWC data - Starting from the dataset that powered ML research for 7 years
  • - Independent - Not owned by Meta, Google, or any big tech
  • - Community-driven - Submit results, report issues, improve accuracy together

The Lesson

Papers with Code's shutdown underscores a risk the academic community faces: critical research infrastructure controlled by commercial entities can disappear without warning. Meta had every right to shut it down. They owned it. That's the problem.

The ML community needs benchmark tracking infrastructure that isn't dependent on corporate goodwill. That's what we're building.

Frequently Asked Questions

Is Papers with Code still working?

No. Papers with Code was shut down by Meta in July 2025 and now redirects to Hugging Face. The SOTA leaderboards and benchmark tracking are gone.

What is the best Papers with Code alternative?

CodeSOTA is building comprehensive ML benchmarks starting with OCR and document AI. Unlike aggregator sites, we verify results ourselves and focus on practical recommendations.

Can I still access Papers with Code data?

Historical data is archived at paperswithcode/paperswithcode-data on GitHub, but it's not being updated. CodeSOTA provides fresh, maintained benchmark data.

Does Hugging Face replace Papers with Code?

Partially. Hugging Face has trending papers and some leaderboards (like Open LLM Leaderboard), but lacks the comprehensive task-based SOTA tracking that Papers with Code provided.

Ready to contribute?

Learn how the data flywheel works and help build the future of ML benchmarks.