Aider Llm Coding Leaderboard Scraper

Gather LLM code-editing benchmark scores directly from Aider's edit and polyglot leaderboards. Select your preferred benchmark mode, adjust the output row limits, and export clean structured datasets in JSON, CSV, or Excel formats effortlessly.

What it does

This Actor extracts leaderboard metrics from Aider's benchmark rankings and outputs each model's evaluation results into structured dataset rows. You can easily analyze pass rates, edit formats, total test counts, and benchmark dates. Results are automatically deduplicated by model name within each run to ensure clean, actionable datasets.

Modes

• Edit leaderboard: Collects code-edit benchmark scores evaluating code modification performance.
• Polyglot leaderboard: Collects multi-language coding scores across diverse programming languages.

Input example

{
  "mode": "edit",
  "maxItems": 50,
  "requestDelaySecs": 1,
  "userAgent": "ActorResearch/1.0",
  "timeoutSecs": 300,
  "requestTimeoutSecs": 30
}

Parameter	Type	Default	Description
mode	string	edit	Choose between edit or polyglot.
maxItems	integer	100	Maximum number of successful rows to return (up to 1000).
requestDelaySecs	number	1	Delay between consecutive requests in seconds.
userAgent	string	ActorResearch/1.0	Custom User-Agent header string.
timeoutSecs	integer	300	Total overall run execution timeout in seconds.
requestTimeoutSecs	integer	30	Individual request timeout in seconds.
proxyConfiguration	object	No proxy	Optional proxy setup. Recommended only if experiencing access blocks.

Output example

{
  "dirname": "2024-05-01-20-05-59--direct-opus-filenames-outside-fence",
  "test_cases": 133,
  "model": "claude-3-opus-20240229",
  "_released": "2024-02-29",
  "edit_format": "diff",
  "commit_hash": "f4b1797-dirty, f4b1797",
  "pass_rate_1": 53.4,
  "pass_rate_2": 68.4,
  "percent_cases_well_formed": 100.0,
  "error_outputs": 2,
  "num_malformed_responses": 0,
  "user_asks": 0,
  "lazy_comments": 0,
  "syntax_errors": 0,
  "indentation_errors": 0,
  "exhausted_context_windows": 0,
  "test_timeouts": 1,
  "command": "aider --opus",
  "date": "2024-05-01",
  "versions": "0.30.2-dev",
  "seconds_per_case": 32.4,
  "total_cost": 13.8395,
  "scraped_at": "2026-10-26T00:00:00+00:00"
}

The overview dataset table highlights key details such as model names, release dates, test cases, pass rates, and edit formats. If an individual request encounters an issue, the Actor records an error item detailing the status and error message without interrupting the rest of your run.

Use cases

• Model comparison: Evaluate and rank AI coding models by their benchmark pass rates without manually parsing webpage layouts.
• Benchmark tracking: Run scheduled extractions to track performance improvements and pass rate changes over time.
• Research & analytics: Pull edit and polyglot evaluation data straight into CSV or Excel files for Jupyter Notebooks or internal reports.
• Tooling optimization: Monitor how edit formats and completion costs evolve across newer model releases.

FAQ

Who is this tool for?

AI researchers, software engineers, and data analysts looking for structured Aider leaderboard benchmarks ready for spreadsheet analysis.

What formats can I export?

You can export datasets in JSON, CSV, Excel, and other standard formats supported by Apify.

Can I run this on a regular schedule?

Yes, you can schedule regular automated runs to keep your dataset updated automatically.

How many rows can one run extract?

A single run can collect up to 1000 items, governed by your maxItems setting.

Trademark Notice

Aider and related names, logos, and trademarks belong to their respective owners. Their use in this documentation is solely to identify the source of the benchmark data and does not imply affiliation, sponsorship, endorsement, or partnership.

Disclaimer

This Actor is an independent tool and is not affiliated with, endorsed by, sponsored by, or otherwise officially associated with Aider or its owners. All data is collected from publicly accessible benchmark resources. Users are responsible for complying with applicable terms of use, rate limits, and licensing requirements.

Is web scraping legal?

This Actor extracts publicly accessible data without bypassing security controls or paywalls. While collecting public data is widely permitted, compliance depends on how data is gathered and used:

• Personal data: When collecting identifiers like names or contact details, ensure compliance with applicable data protection laws such as GDPR or CCPA.
• Copyrighted content: Public text, images, and visual assets are often protected by intellectual property rights. Avoid directly duplicating or redistributing full creative works.
• Platform terms & rate limits: Practice responsible data extraction by avoiding aggressive traffic spikes, adhering to rate limits, and respecting published platform policies.
• Disclaimer: This summary serves informational purposes only and does not constitute formal legal counsel. Seek professional legal advice for specific compliance queries.

For a detailed guide, see Apify's legal overview: Is web scraping legal?.