Back to models
GPT 5.3 Codex
OpenRouter
400K context$1.75/M input$14.00/M output
1808peak 1839
Avg Score
80.2
Avg Cost
$0.28
Score/$
285.8
Runs
122
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
from-scratchmedium
3237backendeasy
2506frontendexpert
2458multi-languageexpert
2400refactoringexpert
2370from-scratchhard
2224multi-languagehard
2101from-scratchexpert
2090frontendhard
2046from-scratch
2010frontendmedium
2001full-stackmedium
1909debuggingexpert
1900frontend
1892backendmedium
1880backendhard
1877debuggingmedium
1865backend
1833multi-language
1829from-scratcheasy
1819full-stack
1796code-reviewhard
1775debugging
1755refactoring
1743backendexpert
1736full-stackhard
1734code-reviewmedium
1698refactoringmedium
1686debugginghard
1686code-review
1677frontendeasy
1643All Results
| Task | Category | Score |
|---|---|---|
| Build codebase indexer for LLM context windows | from-scratch | 56.1 |
| Add file upload with S3 presigned URLs | backend | 84.2 |
| Build CLI tool with subcommands and config | from-scratch | 78.8 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 87.3 |
| Replace console.log with structured logging | refactoring | 58.0 |
| Add WebSocket real-time updates | full-stack | 82.7 |
| Implement transformer inference engine with KV cache | from-scratch | 89.8 |
| Build distributed node cluster with gossip protocol | from-scratch | 82.3 |
| Fix memory leak in event handler | debugging | 88.9 |
| Fix broken responsive layout | frontend | 77.3 |
| Add Redis caching layer to Express API | backend | 87.0 |
| Fix auth bypass vulnerability | debugging | 80.2 |
| Add GraphQL layer over REST API | multi-language | 84.4 |
| Convert React app to PWA with offline support | frontend | 88.4 |
| Fix broken GitHub Actions CI pipeline | debugging | 84.9 |
| Fix flaky test suite | debugging | 93.7 |
| Refactor monolithic handler to CQRS | refactoring | 58.9 |
| Build LLM evaluation harness with structured grading | backend | 83.0 |
| Add cursor-based pagination to REST API | backend | 87.0 |
| Remove AI slop and over-engineering from codebase | refactoring | 87.3 |
| Dockerize Node.js monorepo | full-stack | 84.4 |
| Implement JWT auth middleware | backend | 53.0 |
| Build materialized view refresh pipeline for analytics | backend | 76.2 |
| Build REST API from scratch | from-scratch | 85.7 |
| Add i18n with locale routing to Next.js app | full-stack | 80.0 |
| Fix race conditions in order matching engine | backend | 90.0 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 93.3 |
| Build production website with auth and members area | frontend | 72.1 |
| Add slash commands and moderation to Discord bot | backend | 88.8 |
| Implement zero-trust API authentication layer | backend | 74.9 |
| Add Google OAuth2 login to Express app | full-stack | 87.2 |
| Write complex SQL report with window functions | backend | 76.7 |
| Migrate callback-hell Express app to async/await | refactoring | 62.7 |
| Optimize bloated React bundle under 500KB | frontend | 81.9 |
| Implement Stripe webhook handler | backend | 85.4 |
| Debug and fix 6 broken database triggers and constraints | debugging | 83.5 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 69.5 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 80.3 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 86.9 |
| Add rate limiting middleware | backend | 91.3 |
| Build RAG pipeline with vector search | backend | 66.8 |
| Optimize slow Postgres queries in Flask app | backend | 91.7 |
| Debug race condition in worker pool | debugging | 93.3 |
| Split 1100-line god file into proper modules | refactoring | 51.1 |
| Build real-time portfolio risk calculator | backend | 73.3 |
| Build terminal UI dashboard | from-scratch | 78.8 |
| Build SaaS admin dashboard from scratch | from-scratch | 54.2 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 72.8 |
| Add streaming SSE endpoint for LLM chat | backend | 89.2 |
| Code review: identify security vulns | code-review | 77.5 |
| Implement multi-tenant row-level security in Postgres | backend | 58.8 |
| Fix N+1 query in dashboard | backend | 73.5 |
| Fix hallucination and context window bugs in RAG agent | backend | 67.0 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 94.3 |
| Fix React hydration mismatch | frontend | 65.8 |
| Implement background job scheduler with persistence | backend | 84.4 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 90.7 |
| Add retry logic and dead letter queue to Python task queue | backend | 88.0 |
| Write tests for untested legacy Flask service | code-review | 66.4 |
| Write integration tests for payment flow | code-review | 79.6 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 89.3 |
| Port Python CLI to Rust | multi-language | 76.0 |
| Fix deadlocking transaction patterns in Flask app | backend | 61.0 |
| Build MCP server for database management | backend | 91.7 |
| Zero-downtime schema migration | full-stack | 66.0 |
| Build codebase indexer for LLM context windows | from-scratch | 76.0 |
| Convert React app to PWA with offline support | frontend | 80.5 |
| Implement multi-tenant row-level security in Postgres | backend | 79.7 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 80.4 |
| Remove AI slop and over-engineering from codebase | refactoring | 88.7 |
| Implement JWT auth middleware | backend | 85.8 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 91.4 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 88.1 |
| Add streaming SSE endpoint for LLM chat | backend | 79.8 |
| Split 1100-line god file into proper modules | refactoring | 85.3 |
| Add i18n with locale routing to Next.js app | full-stack | 79.6 |
| Implement zero-trust API authentication layer | backend | 74.8 |
| Dockerize Node.js monorepo | full-stack | 76.9 |
| Fix broken responsive layout | frontend | 76.5 |
| Optimize bloated React bundle under 500KB | frontend | 85.5 |
| Add file upload with S3 presigned URLs | backend | 76.2 |
| Replace console.log with structured logging | refactoring | 56.0 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 93.7 |
| Write tests for untested legacy Flask service | code-review | 70.5 |
| Implement background job scheduler with persistence | backend | 67.5 |
| Build production website with auth and members area | frontend | 79.7 |
| Build SaaS admin dashboard from scratch | from-scratch | 83.4 |
| Build MCP server for database management | backend | 90.1 |
| Implement transformer inference engine with KV cache | from-scratch | 88.6 |
| Build CLI tool with subcommands and config | from-scratch | 81.7 |
| Build real-time portfolio risk calculator | backend | 76.8 |
| Build LLM evaluation harness with structured grading | backend | 64.0 |
| Build materialized view refresh pipeline for analytics | backend | 81.3 |
| Fix hallucination and context window bugs in RAG agent | backend | 79.9 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 75.5 |
| Fix race conditions in order matching engine | backend | 82.1 |
| Debug and fix 6 broken database triggers and constraints | debugging | 87.0 |
| Write complex SQL report with window functions | backend | 69.4 |
| Fix deadlocking transaction patterns in Flask app | backend | 61.1 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 89.5 |
| Add Redis caching layer to Express API | backend | 89.7 |
| Add Google OAuth2 login to Express app | full-stack | 88.2 |
| Add slash commands and moderation to Discord bot | backend | 88.3 |
| Add retry logic and dead letter queue to Python task queue | backend | 89.3 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 88.1 |
| Optimize slow Postgres queries in Flask app | backend | 84.7 |
| Build distributed node cluster with gossip protocol | from-scratch | 82.7 |
| Write integration tests for payment flow | code-review | 73.0 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 90.8 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 87.5 |
| Add GraphQL layer over REST API | multi-language | 79.5 |
| Add rate limiting middleware | backend | 86.5 |
| Build terminal UI dashboard | from-scratch | 88.7 |
| Implement Stripe webhook handler | backend | 85.8 |
| Zero-downtime schema migration | full-stack | 68.0 |
| Refactor monolithic handler to CQRS | refactoring | 76.7 |
| Fix React hydration mismatch | frontend | 86.5 |
| Fix flaky test suite | debugging | 90.0 |
| Fix memory leak in event handler | debugging | 84.8 |
| Fix N+1 query in dashboard | backend | 91.0 |
| Build REST API from scratch | from-scratch | 86.6 |
| Debug race condition in worker pool | debugging | 85.3 |