Back to models
Claude Sonnet 4.5
Anthropic
200K context$3.00/M input$15.00/M output
1625peak 1661
Avg Score
75.5
Avg Cost
$0.25
Score/$
303.6
Runs
62
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
frontendexpert
2388multi-languageexpert
2304code-reviewhard
2170from-scratchexpert
2072frontendhard
1997backendeasy
1923from-scratcheasy
1918refactoringexpert
1912refactoringmedium
1894from-scratchmedium
1870from-scratchhard
1832refactoring
1823from-scratch
1790full-stackmedium
1699backendmedium
1695backendhard
1669code-review
1662backend
1648multi-language
1620code-reviewmedium
1615debuggingexpert
1605backendexpert
1586frontendeasy
1555frontend
1546debugginghard
1529debugging
1521full-stack
1512debuggingmedium
1379frontendmedium
1371full-stackhard
1327multi-languagehard
927All Results
| Task | Category | Score |
|---|---|---|
| Write tests for untested legacy Flask service | code-review | 52.0 |
| Implement background job scheduler with persistence | backend | 52.1 |
| Add Redis caching layer to Express API | backend | 84.6 |
| Add Google OAuth2 login to Express app | full-stack | 60.0 |
| Fix memory leak in event handler | debugging | 81.8 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 76.0 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 56.0 |
| Implement Stripe webhook handler | backend | 85.8 |
| Port Python CLI to Rust | multi-language | 74.0 |
| Build SaaS admin dashboard from scratch | from-scratch | 67.5 |
| Code review: identify security vulns | code-review | 82.6 |
| Add retry logic and dead letter queue to Python task queue | backend | 82.3 |
| Build RAG pipeline with vector search | backend | 65.7 |
| Add file upload with S3 presigned URLs | backend | 82.5 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 86.2 |
| Add streaming SSE endpoint for LLM chat | backend | 57.5 |
| Optimize bloated React bundle under 500KB | frontend | 82.8 |
| Convert React app to PWA with offline support | frontend | 44.7 |
| Add i18n with locale routing to Next.js app | full-stack | 66.2 |
| Replace console.log with structured logging | refactoring | 76.8 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 77.9 |
| Implement multi-tenant row-level security in Postgres | backend | 81.3 |
| Implement JWT auth middleware | backend | 87.3 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 89.0 |
| Remove AI slop and over-engineering from codebase | refactoring | 90.3 |
| Build codebase indexer for LLM context windows | from-scratch | 79.6 |
| Fix broken responsive layout | frontend | 75.3 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 85.5 |
| Split 1100-line god file into proper modules | refactoring | 86.7 |
| Implement zero-trust API authentication layer | backend | 72.4 |
| Dockerize Node.js monorepo | full-stack | 78.3 |
| Build production website with auth and members area | frontend | 76.2 |
| Implement transformer inference engine with KV cache | from-scratch | 87.7 |
| Build CLI tool with subcommands and config | from-scratch | 76.5 |
| Build MCP server for database management | backend | 84.4 |
| Write integration tests for payment flow | code-review | 75.1 |
| Build LLM evaluation harness with structured grading | backend | 74.0 |
| Write complex SQL report with window functions | backend | 83.0 |
| Debug and fix 6 broken database triggers and constraints | debugging | 84.4 |
| Optimize slow Postgres queries in Flask app | backend | 64.5 |
| Build real-time portfolio risk calculator | backend | 63.0 |
| Build materialized view refresh pipeline for analytics | backend | 66.0 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 89.2 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 70.0 |
| Add slash commands and moderation to Discord bot | backend | 77.8 |
| Fix deadlocking transaction patterns in Flask app | backend | 66.7 |
| Fix race conditions in order matching engine | backend | 76.8 |
| Fix hallucination and context window bugs in RAG agent | backend | 77.5 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 87.3 |
| Add GraphQL layer over REST API | multi-language | 52.6 |
| Build distributed node cluster with gossip protocol | from-scratch | 74.1 |
| Fix auth bypass vulnerability | debugging | 93.7 |
| Add rate limiting middleware | backend | 81.5 |
| Zero-downtime schema migration | full-stack | 68.1 |
| Fix flaky test suite | debugging | 72.7 |
| Add cursor-based pagination to REST API | backend | 77.5 |
| Fix N+1 query in dashboard | backend | 88.1 |
| Refactor monolithic handler to CQRS | refactoring | 71.2 |
| Debug race condition in worker pool | debugging | 79.1 |
| Build terminal UI dashboard | from-scratch | 66.3 |
| Build REST API from scratch | from-scratch | 89.5 |
| Fix React hydration mismatch | frontend | 69.2 |