Show HN: Community benchmark for AI coding models with blind voting

1 month ago 1

Community-powered benchmarks from real developers

Upload existing code (refactoring, debugging, code review) → Compare 6 premium AI models → Vote on the winner → Build the world's most accurate benchmark

GPT-5Claude Opus 4.1Claude Sonnet 4.5Grok 4Gemini 2.5 Proo3

Public Beta • 3 Free Evaluations Daily

✓ No credit card required • ✓ Results via email within 24hrs • ✓ 6 top AI models tested

See Who's Winning

Community-driven rankings updated in real-time as developers vote on their favorite solutions

How It Works

See All 6 Models Side-by-Side

Compare outputs, AI judge scores, and community votes in real-time. Every evaluation helps build the world's most accurate benchmark.

Email When Ready

Get notified via email when your evaluation is ready (typically within 24hrs). The current champion model scores all outputs instantly.

You Cast the Final Vote

Pick the winner and explain why in your own words (required comment builds qualitative insights)

Leaderboard Updates Live

Your vote instantly updates win rates and rankings for the entire community to see

Try It Free →

CodeLens.AI Platform - Compare 6 AI Models Side-by-Side

Why Choose CodeLens.AI?

Stop relying on vendor marketing claims. Get unbiased, community-driven insights on which AI model solves YOUR specific coding challenge best. Every evaluation helps build the world's most accurate benchmark of real-world LLM performance.

How It Works

Four simple steps to discover which AI model solves your code challenge best

Submit Your Code Challenge

Paste your code and describe the task. Our AI automatically detects the domain, task type, and language—no dropdowns needed.

All 6 Models Compete

We run your challenge through GPT-5, Claude Opus 4.1, Claude Sonnet 4.5, Grok 4, Gemini 2.5 Pro, and o3 in parallel.

AI Judge + You Vote

The current champion model judges all outputs. Then YOU vote on the winner and share your reasoning (required comment).

Build the Benchmark

Your vote updates the live leaderboard. Over time, we build the world's most accurate dataset of real-world LLM performance.

What Makes Us Different

🎯

Real Tasks

No synthetic benchmarks. Real code from real developers.

🗳️

Community Truth

Developers decide winners, not algorithms or vendors.

🔍

Task-Specific

See which model excels at security vs refactoring vs architecture.

Submit Your First Challenge →

No credit card required • Results via email within 24hrs • 6 top AI models tested

Frequently Asked Questions

Everything you need to know about CodeLens.AI

Ready to Find Your Perfect AI Model?

Discover which AI model solves your code challenges best

FreeNo credit cardResults in 2 minutes

Read Entire Article